NO177603B - Method of Preparation of Hybrid Proteins, DNA Sequence, Hybrid Vector and Eukaryotic Host Cell - Google Patents

Abstract

Novel single-chain hybrid plasminogen activators having an amino acid sequence composed of at least two subsequences corresponding in amino acid identity and number to subsequences of human t-PA and of human u-PA, and mutants thereof in which at least one of the N-glycosylation sites is modified such that glycosylation cannot take place at these sites exhibit valuable pharmacological properties. The hybrid plasminogen activators are produced by recombinant DNA technology.

Description

The present invention relates to a method for the production of a single-chain hybrid plasminogen activator consisting of the following amino acid sequences of human t-PA shown in Figure 1 and human u-PA shown in Figure 3, where the hybrid plasminogen activator is selected from the group consisting of tPA(1 -3)-tPA(176-275)-uPA(159-411), tPA(1-49)-tPA(176-275)-uPA(159-411, and tPA(1-86 )-tPA(176- 275 )-uPA(159-411), DNA sequence, hybrid vector for use in a yeast or mammalian host cell and eukaryotic host cell transformed with a hybrid vector.

Blood clots are the main cause of illness and death in developing countries. Blood clots consist of fibrin which is formed from its soluble precursor fibrinogen with the help of the enzyme thrombin. A host of enzymes and other substances ensure that clots only normally form when and where they are needed to prevent blood loss.

Mammalian plasma contains an enzymatic system, the fibrinolytic system, which can dissolve blood clots. A component of the fibrinolytic system is a group of enzymes called plasminogen activators, which transfer plasminogen (an inactive proenzyme form of plasmin) to the proteolytic enzyme plasmin. Plasmin then degrades the fibrin network of the clumps and forms soluble products. In cases where the thrombolytic capacity of the body is not sufficient for the removal of intravascular thrombi, e.g. in patients suffering from blood clots or post-operative complications, it may be indispensable to use exogenously administered thrombolytic agents.

Two types of plasminogen activators (hereafter referred to as "PAs") can be isolated from human body fluids or cells: urokinase or urokinase-type plasminogen activators (hereafter referred to as "u-PA"), a serine protease that is present e.g. in human urine and kidney cells, and tissue-type plasminogen activator (hereinafter referred to as "t-PA") which is produced by endothelial cells and which is found in a large number of endocrine tissues.

Both t-PA and u-PA exist in two molecular forms: a single-chain form (often referred to as "sc-t-PA" and "sc-u-PA", respectively) and a two-chain (tc) form . The single-chain or pro-enzyme form is converted to a two-chain form by means of proteolytic enzymes at well-defined positions in the polypeptide sequence. The resulting two chains of the processed PA protein remain bound to each other via a sulfur-sulfur bridge. The carboxy-terminal fragment or B-chain carries out the enzymatic activity of PA, while the amino-terminal A-chain contains regulatory units such as fibrin-binding sites. The specific binding of an inactive sc-PA to components of the clot, such as fibrin, followed by its transfer to the active tc-PA by catalytic amounts of proteolytic enzymes present at that site results in an efficient site-specific (site -specific) drug.

t-PA and u-PA are encoded by two different genes, can be distinguished immunologically and enzymatically and have different response profiles to inhibitors, stimulators and activators. Only t-PA is strongly inhibited by the protease inhibitor from Erytrina latissima (DE-3). T-PA activity is strongly stimulated by fibrin and fibrin fragments, while u-PA activity is insensitive to stimulation by fibrin and its fragments. Another feature that distinguishes the two PA enzymes from each other is that tc-t-PA has a high affinity for fibrin and fibrin fragments, while tc-u-PA has no appreciable fibrin affinity.

Due to the unsatisfactory serum stability of injected tPA, the low affinity of tc-u-PA for fibrin and that the fibrin affinity of sc-u-PA is supposed to be indirect, i.e. that it needs an additional blood factor (cf. D.J. Binnema et al., 8th Int. Congress of Fibrinolysis, Vienna, 1986), there is a continuing need for improved plasminogen activators that have a high affinity for fibrin, a more appropriate response to stimulators, a reduced inactivation by inhibitors and a longer effective half-life in the blood circulation.

It is therefore the purpose of this invention to produce new hybrid plasminogen activators which contain the advantageous properties of t-PA and u-PA while they should not contain the undesirable properties of the parent enzymes.

It is surprising that, in the treatment of thrombosis and other conditions where it is desired to produce fibrinolysis through plasminogen activation, the single-chain hybrid PA proteins have much better biological properties when compared to single-chain t-PA and u-PA . More specifically, compared to native PA, smaller quantities of the novel PA molecules are required according to this invention to lyse blood clots in vivo. The single-chain hybrid PA molecules can be produced in large quantities through recombinant DNA technology and when injected into patients will only be converted to their two-chain form by fibrin at the site of the clot to be lysed. Two-chain hybrid PA molecules have been described in the literature (European Patent Application No. 155,387; K.C. Robbins, 8th International Congress of Fibrinolysis, Vienna, 1986), but the more desirable single-chain forms of the hybrid PA molecules cannot be produced at the protein level as it appears from the aforementioned literature, but can only be produced in large quantities and on an industrial scale using recombinant DNA technology.

Accordingly, it is further the purpose of this invention to provide means and methods for the production of said single-chain u-PA/t-PA hybrid proteins. Such agents consist of DNA encoding said u-PA/t-PA hybrid proteins, hybrid vectors containing said DNA and hosts transformed with said hybrid vectors. Methods are also provided for the production of said single-chain u-PA/t-PA hybrid proteins, said DNA, said hybrid vectors and said hosts. This invention also provides a more cost-effective process for the production of two-chain hybrid PA molecules because the single-chain products of the recombinant DNA can be cleaved in vitro by means of appropriate proteolytic enzymes, such as plasmin.

Like other serine proteases involved in the fibrinolytic and coagulation system of the blood, u-PA and t-PA have large non-catalytic segments assembled in chain A linked to the catalytic region (chain B). The non-catalytic A-chain of t-PA can be divided into discrete domains: "finger" domain, "growth factor" domain and two "pretzel" structures while the A-chain of u-PA consists of a "growth factor " domain and a "pretzel" structure [for reference see L. Patthy, Cell 41, 657,663 (1985)]. The catalytic site of the B chains is formed by His, Asp, Ser residues at positions 322, 371 and 478 (t-PA) and 204, 255 and 356 (u-PA), respectively, and is essential for fibrinolytic activity.

A protein domain is a structural and/or functional entity within the entire structure of the entire protein. For example, four domains in the t-PA A chain (finger, growth factor, and two pretzel domains) follow each other in series. The boundaries of the domains are best defined by the positions of the exon-intron boundaries (junctions) in the corresponding DNA sequence (L. Patthy, above). However, for practical reasons, the minimal size of each domain has been defined by the amino acid sequence between the first and last cysteine residues within each domain that are likely to be involved in S-S bridging. Amino acids before and after these cysteine residues from the flanking domains are defined as boundary sequences (J, junction). The positions of the exon-intron boundaries (see above) are within these J regions.

Single-chain t-PA can therefore be represented by the following formula:

T - F - <J>1 - G - J2 - % - J3 - K2 - J4 - TPA<B>

in which T represents the N-terminal part consisting of amino acids 1 to 5, F is finger domain consisting of amino acids 6 to 43, G is growth factor domain consisting of amino acids 51 to 84, Ki is kringle 1 structure consisting of amino acids 92 to 173, K2 is the ring 2 structure consisting of amino acids 180 to 261, TPA<B> is the catalytic serine protease region consisting of amino acids 307 to 527 and J^

(amino acids 44 to 50), J2 (amino acids 85 to 91), J3 (amino acids 174 to 179) and J4 (amino acids 262 to 306) are boundary sequences joining the domain segments.

Single-chain u-PA can be represented using the following formula:

T'-U-J5-K-J6- UPA<B>

where T' represents the N-terminal part consisting of amino acids 1 to 12, U is the growth factor domain consisting of amino acids 13 to 42, K is the pretzel structure consisting of amino acids 50 to 131, UPA<B> is the catalytic serine protease region consisting of amino acids 189 to 411 and J5 (amino acids 43 to 49) and J5 (amino acids 132 to 188) are boundary sequences joining the domain segments.

The J4 and J6 boundary sequences each contain the activation (processing) site and, N-terminal thereto, a cysteine residue involved in a sulfur-sulfur bridge to the catalytic (B-chain) region.

It was surprising to find that single-chain hybrid PA consisting of the catalytic serine protease region of one PA (TPA<B> or UPA<B>) attached to an amino acid sequence consisting of all or discrete A-chain domains of the other PA or discrete domains of both PAs that have valuable pharmacological properties.

The method according to the present invention is therefore characterized by growing under appropriate nutritional conditions a transformed yeast or mammalian host cell containing a DNA coding for said hybrid plasminogen activator and isolating said hybrid plasminogen activator.

The present invention further relates to a DNA sequence characterized in that it consists of the nucleotide sequences coding for a single-chain hybrid plasminogen activator consisting of the following amino acid sequences of human t-PA shown in figure 1 and of human u-PA shown in figure 3, the hybrid plasminogen activator is selected from the group consisting of tPA(1-3)-tPA(176-275)-uPA(159-411), tPA(1-49)-tPA(176-275)-uPA(159-411, and tPA( l-86)-tPA(176-275)-uPA(159-411).

Eybrid vector for use in a yeast or mammalian host cell, characterized in that it comprises a DNA sequence, coding for a single-chain hybrid plasminogen activator consisting of the following amino acid sequences of human t-PA shown in Figure 1 and of human u-PA shown in Figure 3, wherein the hybrid plasminogen activator is selected from the group consisting of tPA(1-3 )-tPA(176-275 )-uPA(159-411), tPA(1-49 )-tPA(176-275)-uPA(159-411 , and tPA(1-86)-tPA(176-275)-uPA(159-41l), the hybrid vector being able to express the given gene, are also described.

The invention further comprises a eukaryotic host cell transformed with a hybrid vector, characterized in that it comprises a DNA sequence coding for a single-chain hybrid plasminogen activator consisting of the following amino acid sequences of human t-PA shown in Figure 1 and human u-PA shown in Figure 3 , the hybrid plasminogen activator being selected from the group consisting of tPA(1-3)-tPA(176-275)-uPA(159-411), tPA(1-49)-tPA(176-275)-uPA(159- 411, and tPA(1-86)-tPA(176-275)-uPA(159-411), the host cell being able to express the given gene.

Preferably, the hybrid PA amino acid sequence begins with the N-terminal sequence of t-PA (t, amino acids 1 to 5) or u-PA (T', amino acids 1 to 12) or begins with any border sequence that is N-terminal linked to the first domain of the hybrid PA or with a fragment of such a border sequence on which the fragment preferably has at least amino acid residues.

In hybrid PA, the A-chain domains are attached via natural boundary sequences (e.g. J^, J2, J3 and J5), compound boundary sequences or hybrid boundary sequences or fragments thereof. Therefore, the first domain of the second domain is at the boundary sequence naturally present at the C-terminus of the first domain, at the boundary sequence naturally present at the N-terminal side of the second domain, at a composite boundary sequence consisting of said joining sequences or of fragments thereof.

The A chain domains of hybrid PA are linked to the B chain serine protease region (TPA<B> or UPA<B>) by means of a boundary sequence selected from the group consisting of the boundary sequence J4 which attaches the A chain to the B chain in human t-PA, the border sequence J(, which attaches A chain to B chain in human u-PA and a hybrid sequence containing the subsequences of said border sequences wherein said border sequence contains a processing site that can be cut by plasmin and, N -terminal thereto, a cysteine residue that can participate in a sulfur-sulfur bridge to the catalytic B-chain region, where the boundary sequence preferably has at least forty and up to sixty amino acid residues.

The boundary of the domains at a position defined by the exon-intron boundaries of the corresponding DNA is preferred. It is preferred that the boundary of A chain to B chain is at the activation site.

In a single chain hybrid plasminogen activator selected from the group consisting of such a hybrid plasminogen activator containing an A chain essentially consisting of the u-PA growth factor domain and the t-PA kringle 2 domain linked in series to the catalytic region (B-chain) of t-PA, and a hybrid plasminogen activator containing an A-chain essentially consisting of the t-PA kringle 2 domain or of the t-PA finger and kringle 2 domains linked in series to the catalytic region (B -chain) to u-PA, in which the boundary between the A-chain domain(s) and B-chain is at the activation site, is particularly preferred.

Below follows the hybrid PA

UPA<A>TPA<B>(BC) = [uPA(1-158)-tPA(276-527)],

UPA<A>TPA<B>(BR) = [uPA(1-131)-tPA(263-527)],

TPA<A>UPA<B>(BC) = [tPA(1-275)-uPA(159-411)],

TPA<A>UPA<B>(BR) = [tPA(1-262)-uPA(132-411)],

UK2TJPAB(BR) = [uPA(l-44)-tPA(176-261)-uPA(134-411)] , FUPA<B>(BC) = [tPA(l-49)-tPA(262-275) -uPA(l59-411)], FUPA<B>(BR) = [tPA(1-49)-uPA(134-411)],

FK2UPA<B>(BC) = [tPA(l-49)-tPA(176-275)-uPA(159-411)], FK2UPA<B>(BR) = [tPA(l-49)-tPA(176 -262)-uPA(132-411)], UK2TPA(BC) = [uPA(1-44)-tPA(176-527)],

K2UPA<B>(BC) = [tPA(l-3)-tPA(176-275 )-uPA(159-411)], FGK2UPA<B>(BC) = [tPA(l-86)-tPA(176 -275)-uPA(159-411)] and FGK2UPA<B>(BR) = [tPA(1-86)-tPA(176-262)-uPA(132-411)], where UPA<A> is A -chain of u-PA, TPA<A> is in A chain of t-PA, UPA<B> is B chain of u-PA, TPA<B> is B chain of t-PA, U refers to the growth factor domain of u-PA, K2 refers to the pretzel 2 domain of t-PA, F refers to the finger domain of t-PA, G refers to the growth factor domain of t-PA, (BC) indicates that the boundary between the A-chain domain(s) and the B-chain is at the activation site and (BR) indicates that the A-chain domain(s) is connected to the B-chain via the border sequence naturally connected to the B-chain including the activation site and, N-terminal thereto, the cysteine residue involved in a sulfur-sulfur bridge to the B chain. The numbers refer to the amino acid sequences from u-PA and t-PA, respectively. For example, UK2UPA<B>(BR) = [uPA(1-44 )-tPA-(176-261 )-uPA(134-411)] designates a single-chain hybrid plasminogen activator consisting of amino acids 1-44 (growth factor domain, U) of u-PA and amino acids 176-261 (pretzel 2 domain, K2) of t-PA linked in a linear fashion to amino acids 134-411 (B-chain, UPA<B>) of u-PA.

Hybrid plasminogen activators TPA<A>UPA<B>(BC), FUPA<B>(BC), FGK2UPAB(BC) and, in particular, UK2TPA<B>(BC), FK2UPA<B>(BC) and K2UPA<B >(BC), is particularly preferable.

It is already known that a prerequisite for N-linked glycosylation in mammalian cells is the presence of the tripeptide sequence -Asn-L-Ser(or Thr)- in which Asn is the acceptor and L can be any of the 20 genetically encoded amino acids except for proline or aspartic acid which prevents glycosylation. There are three sites for N-glycosidic bonding in the t-PA molecule (the numbers refer to the position of Asn in the amino acid sequence of t-PA, cf. Fig. 1 to the attached drawings): - Asn117-Ser-Ser- (present in pretzel 1) , Asn<184->Gly-Ser- (present in pretzel 2), and Asn<448->Arg-Thr (present in t-PA B chain). The unique N-linked glycosylation site of u-PA is in the B chain (Asn<302->Ser-Thr, cf. Fig. 3). It is obvious that hybrid PA containing t-PA pretzel 1, t-PA pretzel 2, B-chain of t-PA and/or B-chain of u-PA also includes the respective N-linked glycosylation sites.

To prevent glycosylation at individual (one or more of the) N-glycosylation sites, the tripeptide sequences that are recognized as signals for N-glycosylation must be altered. "Replacement of the Asn and/or Ser (or Thr) residues in the above-mentioned tripeptide sequence with any other amino acid would prevent the formation of glycosidic bonds at these sites. Modification of the N-glycosylation sites is not performed at the protein level. It is an advantage instead of modifying the gene encoding the hybrid PA in such a way that during the expression of said modified gene by a host, a mutant hybrid PA is produced in which one or more of the N-glycosylation sites are altered in such a way that glycosylation cannot take place at these sites It is most convenient to modify all one N-glycosylation sites present in the hybrid PAs according to the invention.

In particular, asparagine is substituted with valine, leucine, isoleucine, alanine or, in particular, glutamine, and serine or threonine with valine, methionine or, in particular, alanine.

Modified hybrid PAs are particularly preferred.

FUPA<B>(Gln302)(BC) = [tPA(1-49 )-tPA(262-275)-uPA(159-301, Gln, 303-411 )],

FK2(Alal86)UPA<B>(Gln302)(BC) = [tpa(1-49)-tPA(176-185, Ala, 187-275)-uPA(159-301, Gln, 303-411)],

UK2(Alal86)TPA<B>(Ala450)(BC) = [uPA(1-44)-tPA(176-185, Ala, 187-449, Ala, 451-527)],

K2(Alal86)UPA<B>(Gln302)(BC) = [tPA(1-3)-tPA(176-185, Ala, 187-275)-uPA(159-301, Glin, 303-411)],

FGK2(Alal86)UPA<B>(Gln302)(BC) = [tPA(1-86)-tPA(176-185), Ala, 187-275)-uPA(159-301, Gln, 303-411)] ,

and further

FK2UPA<B>(Gln302)(BC) = [tPA(1-49)-tPA(176-275)-uPA(159-301, Gln, 303-411 )],

K2UPA<B>(Gln302)(BC) = [tPA(1-3)-tPA(176-275)-uPA(159-301, Gln, 303-411)] ,

UK2TPA<B>(Ala450)(BC) = [uPA(1-44)-tPA(176-449, Ala, 451-527)], and

FGK2TJPA<B>(Gln302)(BC) = [tPA( 1-86 )-tPA( 176-275 )-uPA( 159-301, Gln, 303-411 )].

Hybrid PAs and mutants thereof can be produced by means of recombinant DNA technology consisting, for example, of culturing a transformed host expressing the hybrid PA protein or mutant thereof under conditions permitting expression thereof and isolating the hybrid PA- the protein and mutant hybrid PA protein. More specifically, the desired compounds are prepared by

a) production of DNA encoding a hybrid PA protein or a mutant thereof or synthesizing such DNA chemically, b) incorporation of DNA into an appropriate expression vector, c) transfer of the obtained hybrid vector into a recipient host, d) selection of transformed hosts from untransformed hosts, e.g. by cultivation of by conditions where only

the transformed host survives,

e) culturing the transformed host under conditions allowing expression of the hybrid PA protein, and f) isolating the hybrid PA protein or mutants thereof.

The steps involved in the production of the hybrid PA proteins using recombinant DNA technology will be discussed in more detail below.

DNA encoding hybrid PA proteins

DNA according to the invention preferably has flanking sequences at the terminal end. In particular, these flanking sequences contain appropriate restriction sites that allow integration of the DNA into appropriate vectors.

Furthermore, the DNA according to the invention contains the signal sequence of u-PA or t-PA which is attached to the first codon of the finished hybrid PA coding sequence. When expressed in yeast cells, the DNA according to the invention contains a yeast signal sequence, such as the signal sequence which is naturally bound to the yeast promoter used, especially PH05 or invertase signal sequence.

The nucleotide sequences of the DNA subsequences are preferably similar to the nucleotide sequences found in u-PA cDNA and t-PA cDNA, respectively. However, due to the degeneracy of the genetic code, the nucleotide sequences may differ if the resulting amino acid subsequences remain unchanged. In the DNA encoding a mutant hybrid PA, at least one codon encoding an amino acid essential for N-glycosylation of the hybrid PA protein may be replaced by another codon encoding another amino acid that prevents N-glycosylation of the recognition seat.

The nucleotide sequence of u-PA cDNA and t-PA cDNA is known [cf. W. E. Holmes et al., Biotechnology 3, 923-929 (1985); D. Pennica et al., Nature 301, 214-221 (1984)]. Furthermore, the entire nucleotide sequences of the genomic u-PA and t-PA genes including all introns and extrons are known [cf. A. Riccio et al., Nucl. Acids Res. 13, 2759-2771 (1985); S. J. Friezner-Degen et al., J. Biol. Chem. 261, 6972-6985 (1986)].

Since the cDNA and genome DNA sequences of u-PA and t-PA are known, DNA according to the invention can be made using methods that are known in the field. Methods for preparing these DNAs include chemical synthesis of DNA or preparation of fragments encoding the polynucleotide subsequences of u-PA cDNA and t-PA cDNA and placing them in the most convenient predetermined order including one or more, such as two or three, mutation step.

DNA encoding the mutant hybrid PA can be prepared using methods known in the art. The methods for producing these DNAs include cutting out a portion of the DNA containing the codon for the unwanted amino acid residue from the parent hybrid PA gene and replacing it with a DNA segment in which said codon has been substituted with a deoxyribonucleotide triplet that codes for the desired amino acid residue, or perform the deoxyribonucleotide substitution by means of site-specific mutagenesis.

Chemical synthesis of DNA is well known and uses conventional techniques. Appropriate techniques have been developed by S.A. Narang [Tetrahedron 39, 3 (1983)]. In particular, the methods described in European Patent Application No. 146,785 may be used and are incorporated herein by reference.

Another method for the synthesis of DNA consists of cutting out suitable restriction fragments that code for the polynucleotide subsequences of u-PA and t-PA from u-PA cDNA and t-PA cDNA (or genomic u-PA DNA or t-PA DNA) and use of these fragments for the production of the entire structural gene of hybrid PA. Two strategies can be used. With each of the strategies, care must be taken that the fusion of the fragments occurs at the sites between the domains to keep them intact. The first strategy makes use of suitable restriction seats. When an appropriate restriction site is present at the predetermined boundary site(s) in both u-PA and t-PA DNA, the DNA is cut with the corresponding restriction endonuclease and the fragments are joined by blunt-end or overhanging ligation (depending on which restriction endonuclease) depending on which restriction endonuclease is selected). Alternatively, useful restriction sites can be introduced by, for example, site-specific mutagenesis

[cf. M. J. Zoller et al., Methods Enzymol 100, 468 (1983)]

when care is taken that the mutated DNA does not result in a different amino acid sequence. Natural or artificially introduced restriction seats which are particularly preferred

are those that separate the DNA encoding the A and B chains or the DNA encoding the discrete domains that are in the A chains. In this way, hybrid DNA can be produced encoding hybrid PAs with the desired boundary between the A-chain domains and the serine protease catalytic region. The second strategy comes from the hypothesis that the domain boundaries are best defined using the positions of the exon-intron boundaries in the genomic DNA [cf. L. Patthy, Cell 41, 657-663 (1985)], i.e. positions in cDNA where introns have been spliced. As these positions rarely coincide with the restriction sites, a system has been developed that can be followed in any new construction: in the first step, appropriate restriction fragments coding for the specific domain(s) but also containing additional DNA sequences are that extend beyond the expected fusion point (up to several hundred base pairs) are ligated and subcloned into bacteriophage ml3. In the second step, the DNA sequences that are in excess are cut out by means of in vitro mutagenesis (Zoller et al., supra). This method allows precise fusions that are the correct reading frame at any predetermined nucleotide position and is therefore preferred.

For the production of mutant hybrid PAs, part of the finished hybrid DNA can be cut away using restriction enzymes. A prerequisite for this method is that there are suitable restriction sites in the vicinity of the codon to be changed. A small restriction fragment containing the codon for the unwanted amino acid is removed by endonuclease cutting. A corresponding double-stranded DNA sequence is produced, for example by chemical synthesis, where triplets coding for the desired amino acid are used. The DNA fragment is ligated in the correct orientation to the remaining large fragment, leading to a double-stranded DNA sequence encoding a mutant hybrid. To make it convenient and to facilitate handling of the hybrid gene, it is inserted into a larger DNA segment that has appropriate end nucleotides (link-ers) that allow insertion and cloning of the segment into a cloning vector.

The production of DNA encoding a mutant hybrid PA is carried out by means of site-specific mutagenesis. This method is an in vitro mutagenesis procedure where a defined site within a region of cloned DNA can be changed (cf. the review articles by M.J. Zoller and M. Smith, Methods Enzymol. 100, 468 (1983); D. Botstein and D. Shortle , Science 229, 1193 (1985)]. Mutagenesis can either be performed on the entire hybrid PA gene or on functional portions thereof containing the codon for the undesired amino acid(s). After mutagenesis, the mutated functional portion is ligated to the other portions of the hybrid PA to obtain the mutant hybrid

ON.

The method for mutating the hybrid PA gene or functional parts thereof is characterized in that the single-chain gene or a single-chain DNA containing the PA gene or parts thereof is hybridized to an oligodeoxyribonucleotide primer that is complementary to the region in the the hybrid gene to be mutated except for the mismatch(s) driving the mutation, the hybridized oligodeoxyribonucleotide is used as a primer to initiate the synthesis of the complementary DNA strand, the resulting (partial) double-stranded DNA is transformed into in recipient microorganism strain, the microorganism strain is grown and transformants containing DNA with the modified (mutant) hybrid PA gene are selected.

Hybrid vectors containing hybrid PA DNA

The invention relates to hybrid vectors.

The vector is selected depending on which host cells are intended to be used for the transformation. In principle, any vector that replicates and expresses the desired polypeptide gene of the invention in the chosen host is suitable. Examples of suitable hosts are eukaryotes, which lack or have few restriction enzymes or modifying enzymes, such as yeast, for example Saccharomyces cerevisiae, for example S. cerevisiae GRF18, and further mammalian cells, especially known human or animal cell lines, e.g. myeloma cells, human embryonic lung fibroblasts L-132, COS cells, LTK cells, human malignant melanoma Bowes cells, HeLa cells, SV-40 virus-transformed African green monkey kidney cells COS-7 or Chinese hamster ovary (CHO) cells and variants thereof. The above-mentioned mammalian cells and strains and Saccharomyces cerevisiae, for example S. cerevisiae GRF18, are preferred as the host microorganism.

a. Vectors for use in yeast

Vectors suitable for replication and expression in yeast contain a yeast origin of replication and a yeast selectable genetic marker. Hybrid vectors containing a yeast origin of replication, such as the chromosomal autonomously replicating segment (ars), are retained extrachromosomally within the yeast cell after transformation and are replicated autonomously. Furthermore, hybrid vectors containing sequences homologous to yeast 2jj plasmid DNA can be used. Such hybrid vectors will be integrated by recombination into the 2p plasmids that already exist inside the cell, or replicate autonomously. The 2jj sequences are particularly suitable for plasmids with a high transformation frequency and a high copy number.

Suitable marker genes for yeast are, in particular, those that confer antibiotic resistance to the host or, in the case of auxotropic yeast mutants, genes that complement host lesions. Corresponding genes cause, for example, resistance to the antibiotic G418 or lead to prototropy in an auxotropic yeast mutant, for example the URA3, LEU2, HIS3 or TRP1 gene. Hybrid yeast vectors preferably further contain an origin of replication and a marker gene for a bacterial host, especially E. coli, so that the construction and cloning of the hybrid vectors and their intermediates can take place in a bacterial host.

Expression control sequences suitable for expression in yeast are, for example, those derived from yeast genes that are well expressed. Therefore, the promoters of the TRP1 gene, the ADHI or ADHII gene, the acid phosphatase (PH03 or PH05) genes, the isocytochrome gene or a promoter of the glycolysis genes, such as the promoter of enolase, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 3-phosphoglycerate -kinase (PGK), hexokinase, pyruvate decarboxylase, phosphofructokinase, glucose-6-phosphate isomerase, 3-phosphoglycerate mutase, pyruvate kinase, triosephosphate isomerase, phosphoglucose isomerase, invertase and glucokinase genes, can get used. Vectors which are preferred according to this invention contain promoters with the transcriptional control, e.g. the promoters of the PH05 and ADH II genes, which can be turned on or off by varying the growth conditions. For example, the PH05 promoter can be repressed or derepressed simply by increasing or decreasing the concentration of inorganic phosphate in the medium.

The yeast hybrid vectors preferably also consist of the 3' flanking sequence of a yeast gene containing the correct signals for transcription termination and polyadenylation. Suitable 3' flanking sequences are, for example, those of the gene naturally linked to the promoter being used, such as the 3' flanking sequence of the yeast PH05 gene.

b. Vectors for use in mammalian cells

Vectors for replication and expression in mammalian cells often contain DNA of viral origin, e.g. from simian virus 40 (SV 40), Rous sarcoma virus (RSV), adenovirus 2, bovine papilloma virus (BVP), papovavirus BK mutant (BKV), or mouse or human cytomegalovirus (MSMV and HCMV, respectively).

Expression control sequences suitable for use in mammalian cells include, among others, the early and late promoters of SV40, the major late promoter of adenovirus, the promoter of the murine metallothionein gene, and the enhancer-promoter region of mouse or human cytomegalovirus major immediate- early gene, human immunoglobulin enhancer-promoter region, human α-globin promoter combined with the SV40 enhancer and promoters derived from the heat shock genes.

Suitable marker genes for mammalian cells are, for example, the neo and ble genes from transposon Tn5 which confer resistance to the antibiotic G418 and to bleomycin-type antibiotics, respectively, the E. coli gene for hygromycin-B resistance, the dihydrofolate reductase gene (dhfr) from mammalian cells or E. coli which change the phenotype of DHFR" cells to DHFR<+> cells and/or lead to resistance to methotrexa, and the thymidine kinase gene of herpes simplex virus which makes TK" cells phenotypically TK<+> cells.

The hybrid vectors for the mammalian cells preferably contain the 3' untranslated region of a mammalian gene containing signals for proper transcription termination and polyadenylation, such as, for example, the 3' flanking regions of the p<->globin gene. Advantageously, the regions flanking the polypeptide coding region contain one or more native introns having appropriate splice signals at their terminal ends. Such additions are considered necessary as cDNA and prokaryotic DNA such as the above-mentioned selection genes generally lack such transcription and processing signals.

Preferably, such vectors contain an origin of replication and an antibiotic resistance gene for propagation in E.coli. A mammalian origin of replication that is formed either by including in the construction of the vector a eukaryotic origin, such as that derived from SV40 or from another viral source, or may be formed from the host cell chromosome by integration of the vector into the host cell chromosome.

The preferred hybrid vectors for use in mammalian cells include hybrid PA or mutant hybrid PA cDNA flanked upstream of the murine cytomegalovirus major immediate-early gene enhancer-promoter and downstream of the 3' end of the rabbit beta-globin gene, which includes the second intron with its appropriate splicing signals and a polyadenylation sequence. They further contain the sequences encoding the neomycin resistance gene from transposon Tn5 or more optimally from Tn9 or the sequences encoding hygromycin phosphotransferase flanked on its upstream side sequentially by the SV40 virus early promoter which also includes the SV40 origin of replication and the natural promoter of the Tn5 neogene, and on its downstream side a segment of the SV40 early gene that includes the small t-antigen splicing and polyadenylation signals. The entire construct is cloned into a fragment of E.coli plasmid pBR322, which contains the plasmid origin of replication, the ampicillin resistance gene, but lacks so-called poison sequences that inhibit SV40-style DNA replication in mammalian cells. For optimality, a gene encoding dihydrofolate reductase (DHFR) is inserted into the vector, preferably the modular DHFR gene described by R.J. Kaufman et al., [Mol. Cell. Biol. 2, 1304-1319 (1982)] is used. This modular DHFR gene consists, sequentially, of the adenovirus type 2 core promoter, a fragment of an immunoglobulin gene, the coding part of a murine DHFR cDNA and the SV40 early polyadenylation site.

The new preferred hybrid vectors for use in mammalian cells represent an advance in this area. They are superior to prior art vectors in that they contain the strong expression signals for the cloned cDNA located in the mouse cytomegalovirus immediate-early promoter/enhancer and in beta-globin splicing/poly-adenylation sequences in an environment that allows expression on a high level in large numbers of different vertebrate cell types. More specifically, the vectors can be used (a) to transiently express cDNA in normal, i.e. not SV40-transformed, tissue culture cell lines, but (b) even better high copy number in primate cells expressing SV40 T antigen, thus allowing the vector to replicate via its SV40 origin of replication, but also (c) to express such cloned cDNA stably in normal tissue culture cell lines, where the vector can integrate into the host cell chromosome and (d) even better, because of the high copy number, when the vector is introduced into SV40 Tr-ant in gene producing primate cell lines, where the vector can replicate episomally.

For example, the enhancer-promoter region of MCMV contains a DNA starting at nucleotides -835 to -443 and ending at nucleotide +50 (counting from the mENA start) to the 5' region of the MCMV major immediate-early gene. The enhancer-promoter region preferred to MCMV consists of nucleotides -542 to +50.

The 3'-flanking region of the rabbit p<->globin gene consists of the other half of the rabbit beta-globin gene [P. Dierks et al., Proe. Nati. Acad. Pollock. USA 78, 1411-1415 (1981); A. van Ooyen et al., Science 206, 337-344 (1979)] and begins in the second exon, preferably at the BamHI site, thus including the second intron with the signals for splicing at its flanking sequences, and terminating after the polyadenylation signals, preferably at the Bglll site which lies 1.2 kb behind the above-mentioned BamHI site.

The origin of replication of SV40 lies, for example, in the HindIII-SphI fragment of the viral DNA [nucleotides 5171 to 128, origin = position 1; Tooze J. (ed.) DNA Tumor Viruses, Part 2, 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor N.Y. 1982]. Most suitable is the HindIII-Hpall fragment (nucleotides 5171 to 346), which in addition to the origin of replication also contains the viral early enhancer/promoter which promotes transcription of the selected gene into the vector.

The neomycin gene is cloned from a promoter active in tissue culture cells, preferably the SV40 early promoter located on the Hpall-Hindlll fragment mentioned above. The coding sequences of the neomycin gene reside, for example, on a BglII-SmaI fragment from transposon Tn5 [E. Beck et al., Gene 19, 327-336 (1982); P. Southern et al., J. Mol. Appl. The gene. 1, 327-341 (1982); F. Colbere-Garapin et al., J. Mol. Biol. 150, 1-14 (1981)]. It is preferable to add a promoter to the neomycin gene which allows transcription also in E.coli. For example, the natural promoter of the Tn5 neomycin gene may preferably have a Hindi 11-Bgl 11 fragment placed behind the eukaryotic promoter in front of neo-coding sequences (Southern et al., supra) or further upstream in front of the eukaryotic promoter (Colbére -Garapin et al., supra). To be expressed in tissue culture cells, the bacterial neogene must be followed by a polyadenylation signal, preferably a part of the SV40 t antigen gene that also contains splicing signals. The coding sequence of neomycin phosphortransferase, especially the Bglll-SmaI part of the Tn5 fragment mentioned above, can also be replaced that the coding sequence of hygromycin B phosphortransferase preferably in the form of the BamHI fragment of plasmid pLG89 [L. Gritz et al., Gene 25, 179-188 (1983)], which can conveniently be inserted into pSVd [Luedin et al., EMBO-J. 6, 109-114 (1987)], a derivative of pSV2911neo in which a BglII linker has been introduced at the SmaI site in the vector.

Another preferred selection gene uses the coding sequence of the enzyme dihydrofolate reductase, such as in pSV2dhfr (ATCC 37145), which allows not only selection of transformed cell lines but also amplification of the plasmid-associated DNA sequence, often with a proportional increase of the production of plasmid-encoded proteins according to the invention.

The fragment derived from E.coli plasmid pBR322 contains the pBR322 origin of replication and the ampicillin resistance gene. The fragment is preferably taken from a pBR322 derivative, such as pSVOd [P. Mellon et al., Cell 27, 279-288 (1981)] where the so-called poison sequence which would have inhibited the SV40 T antigen-driven replication of the vector has been removed.

In a preparation which is preferred, hybrid vectors are used which have the possibility of replication and phenotypic selection in a eukaryotic host strain consisting of a promoter and a DNA which codes for a hybrid PA or mutant hybrid PA, where said DNA is placed together with the start of transcription and termination signals and translation start and stop signals in said hybrid vector under the control of said promoter so that production of the protein is expressed in a transformed host.

Hybrid vectors according to this invention are produced by methods known in the art, for example by linking the DNA segments containing the promoter, the hybrid PA or mutant hybrid PA coding region, the 3'-flanking sequence and vector DNA.

Different techniques can be used to link DNA segments in vitro. Blunt ends (completely base-paired DNA duplexes) produced by certain restriction endonucleases can be ligated directly with T4 DNA ligase. It is more common to link DNA segments using their single-stranded cohesive ends and are covalently closed using a DNA ligase, e.g. T4 DNA ligase. Such single-stranded "cohesive termini" can be formed by cutting DNA with another class of endonucleases that form overhangs (the two strands of the DNA duplex are cut at different places with a few nucleotides in between). Single chains can also be formed by adding nucleotides to blunt ends or overhanging ends using terminal transferase ("homopolymeric tailing") or by "eating up" a chain in a DNA segment that has a blunt end using a suitable exonuclease, such as X exonuclease. Another method for producing overhanging ends consists in ligating a chemically synthesized end-nucleotide DNA containing a recognition site for an overhanging end-forming endonuclease to a DNA segment having blunt ends and cutting the resulting DNA with the respective endonuclease . The components of the hybrid vectors of this invention are joined together in a predetermined order to ensure proper function.

Hosts transformed with h<y>bride vectors containing hvbrid PA DNA

Another aspect of this invention relates to eukaryotic host organisms transformed with hybrid vectors containing a DNA encoding a hybrid PA consisting of at least two subsequences corresponding in amino acid identity and number to subsequences of human u-PA and human t-PA or which encodes a mutant thereof, and mutants of said host, and processes for the preparation thereof.

Examples of suitable eukaryotic hosts are those mentioned above, particularly yeast strains and mammalian cells. Mutants to transformed host organisms particularly include mutants that have poor hybrid PA or mutant hybrid PA degrading proteases and give greater yields in hybrid PA and mutant hybrid PA, respectively.

The method for producing the transformed eukaryotic hosts consists of transforming or transfecting a eukaryotic host with an expression vector containing a DNA according to this invention and which is regulated by an expression control sequence.

Transformation of the eukaryotic host cell is carried out using methods known in the art. For example, transformation of yeast with the hybrid vectors can be performed according to the method described by Hinnen et al [Proe. Nati. Acad. Pollock. USA 75, 1919(1978)]. This method can be divided into three steps:

(1) Removal of the yeast cell wall or parts thereof.

(2) Treatment of the "naked" yeast cells (spheroplasts) with the transforming DNA in the presence of PEG (polyethylene glycol) and Ca<2+> ions. (3) Regeneration of the cell wall and selection of the transformed cells in a solid agar layer.

Preferred methods:

ad (1): The yeast cell wall is removed enzymatically using various preparations of glucosidases, such as snail intestinal juices (e.g. Glusulase® or Helicase®) or enzyme mixtures from microorganisms (e.g. Zymolyase®) in osmotically stabilized solutions ( eg 1 M sorbitol).

ad (2): The yeast spheroplasts aggregate in the presence of PEG and local fusions of the cytoplasmic membranes are induced. The formation of "fusion-like" conditions is absolutely essential and many transformed yeast cells become diploid or even triploid during the transformation process. Procedures that allow selection of fused spheroplasts can be used to enrich for transformants, i.e. transformed cells can easily distinguish among preselected fusion products.

ad (3): As yeast cells without a cell wall do not divide, the cell wall must be regenerated. This regeneration is done by mixing the spheroplasts into agar. For example, molten agar (about 50°C) is mixed with the spheroplasts. On cooling the solution to yeast growth temperatures (about 30°C), a solid layer is formed. This agar layer is used to

prevent rapid diffusion and loss of essential macromolecules from the spheroplasts and thus facilitate regeneration of the cell wall. Cell wall regeneration, on the other hand, can also be carried out (but with lower efficiency) by plating out the spheroplasts on the surface of ready-made agar layers.

The agar for regeneration is prepared in a way that allows both regeneration and selection of transformed cells at the same time. Since yeast genes encoding enzymes for amino acid biosynthetic pathways are often used as selective markers (supra), the regeneration is preferably carried out in yeast minimal medium agar. If a high yield of the regeneration is required, the following two-step procedures are advantageous (1): cell wall regeneration in a rich complex medium, and (2) selection of the transformed cells by replica-plating the cell layer on selective agar plates.

The introduction of hybrid vectors into mammalian cells is done by transfection in the presence of auxiliary compounds, e.g. diethylaminoethyldextran, dimethyl sulfoxide, glycerol, polyethylene glycol or the like, or as co-precipitates of vector DNA and calcium phosphate. Other suitable methods consist of microinjecting vector DNA directly into the cell nucleus and electroporation, i.e. introduction of DNA by a short electrical pulse that increases the permeability of cell membranes. The subsequent selection of transfected cells can be done using a selection marker that is either covalently integrated into the expression vector or added in separate form. Markers for selection consist of genes that cause resistance to antibiotics or genes that complement a genetic deficiency in the host cell (supra). A system of selection that is preferred uses cells lacking dihydrofolate reductase (DHFR"), eg CHO cells, which absolutely require thymidine, glycine and purines for growth unless an exogenous DHFR gene is added. When a vector containing the hybrid PA gene and in addition a DHFR gene into appropriate DHFR" cells is added, e.g. CHO cells, the transformed cells are selected by increasing the concentration of the anti-folate drug methotrexate to the medium.

A method of selection which is particularly preferred is a method in which suitable mammalian cells, e.g. CHO cells, are treated with co-precipitates of vector DNA containing the hybrid PA gene and a gene encoding antibiotic resistance, e.g. resistance to G-418, and calcium phosphate. The transformed cells are selected by cultivation in the presence of the corresponding antibiotics, e.g. G-418, and/or screening hybrid PA expression.

Transformed host organisms according to the invention can be improved when it comes to the production of hybrid PA or mutant hybrids PA by mutation and selection using methods known in the art. The mutation can be formed, for example, by U.V. radiation or by means of suitable chemicals. Production of protease-deficient mutants is particularly preferred, especially yeast mutants, to avoid proteolytic degradation of the formed hybrid PA and mutant hybrid PA, respectively.

Cultivation of transformed host cells

The invention further contains a method for the production of single-chain hybrid PA which has an amino acid sequence consisting of at least two subsequences which correspond in amino acid identity and number to subsequences of human t-PA and of human u-PA, or mutants thereof, and cultivation by suitable nurturing conditions a transformed eukaryotic host containing a DNA sequence encoding said hybrid PA or mutant hybrid PA and isolating said hybrid PA or mutant thereof.

The transformed host cells are cultured by methods known in the art in a liquid medium containing digestible sources of carbon, nitrogen and inorganic salts.

Different carbon sources can be used for growing the transformed yeast cells according to this invention. Examples of carbon sources which are preferred are digestible carbohydrates, such as glucose, maltose, mannitol or lactose, or an acetate, which can be used either alone or in suitable mixtures. Examples of suitable nitrogen sources are amino acids, such as casamino acids, peptides and proteins and their degradation products, such as tryptone, peptone or meat extracts, yeast extracts, malt extract and also ammonium salts, for example ammonium chloride, sulphate or nitrate, which can either be used alone or in suitable mixtures. Inorganic salts which can also be used are, for example, sulphates, chlorides, phosphates and carbonates of sodium, potassium, magnesium and calcium.

The medium further contains, for example, growth-promoting substances, such as trace elements, for example iron, zinc, manganese and the like, and preferably substances that exert a selection pressure and prevent the growth of cells that have lost the plasmid expression. Therefore, for example, if a yeast strain that is auxotropic in, for example, an essential amino acid is used as the host microorganism, the plasmid preferably contains a gene encoding an enzyme that complements the defect in the host. Cultivation of the yeast strain is carried out in a minimal medium lacking the aforementioned amino acid.

Cultivation is carried out using methods that are known in the field. Conditions for cultivation, pH values of the medium and fermentation time are chosen so as to achieve the highest possible titer of the PA proteins in this invention. Thus, the yeast strain is preferably grown under aerobic conditions with flooding of the culture with shaking or shaking at a temperature of about 20 to 40°C, preferably about 30°C, and a pH value of 5 to 8, preferably at about pH 7 , for about 4 to 30 hours, preferably until a maximum yield of the proteins of this invention is obtained.

Mammalian cells are grown under tissue culture conditions using commercial media supplemented with growth-promoting substances and/or mammalian serum. The cells grow either attached to a fixed support, e.g. a microcarrier or porous glass fibers, or free-flowing in appropriate culture containers. The culture medium is selected in such a way that a selection pressure is exerted and that only those cells survive which still contain the hybrid vector DNA containing the genetic marker. Therefore, for example, an antibiotic is added to the medium when the hybrid vector contains the corresponding antibiotic resistance gene.

When the cell density has reached a high enough value, cultivation is interrupted and the protein is isolated. When using mammalian cells, hybrid PA or mutant hybrid PA protein is usually secreted into the medium. The medium containing the product is separated from the cells, where fresh medium can again be added and used for continuous production. When yeast cells are used, the protein can also accumulate inside the cells, especially in the periplasmic area. In the latter case, the first step for PA protein recovery consists in releasing the proteins from the cells. In most methods, the cell wall is first removed by enzymatic cutting of the cell wall with glucosidases (supra). Alternatively, the cell wall is removed by treatment with chemicals, i.e. thiol reagents or EDTA, which cause the cell wall to be damaged and allow the produced hybrid PA or mutant thereof to be secreted. The resulting mixture is enriched for hybrid PA or for its mutant by conventional methods, such as removal of most of the non-protein substances by treatment with polyethyleneimine, precipitation of the proteins by means of ammonium sulfate, gel electrophoresis, dialysis, chromatography, for example, ion - exchange chromatography, size exclusion chromatography, HPLC or reverse phase HPLC, gel filtration on a suitable Sephadex column, or the like. The final purification of the pre-purified product is achieved, for example, by means of affinity chromatography, for example antibody affinity chromatography, in particular monoclonal antibody affinity chromatography by means of monoclonal anti-t-PA or anti-u-PA antibodies fixed on a non-soluble matrix using methods known in the art or, in the case of hybrid PA containing the catalytic B chain of t-PA, DE-3 affinity chromatography (DE-3 is a protease inhibitor isolated from Erytrina latissima), and the like.

Hybridoma cell lines that produce monoclonal antibodies directed against specific domains of t-PA or u-PA and said monoclonal antibodies are also part of the invention.

For the appropriate preparation of the single-chain form of the hybrid PA or mutant hybrid PA which is substantially free of the two-chain form, a protease inhibitor, such as aprotinin (Trasylol) or basic pancreatic trypsin inhibitor, is advantageously added in during the purification procedure to inhibit traces of proteases which may be present in the culture medium and which may lead to (partial) transfer of the single-chain form to the two-chain form. The final purification is then achieved by chromatography and a column containing a selective affinity reagent.

5. Pharmaceutical preparations

The novel single-chain hybrid PA proteins and mutants thereof produced according to this invention exhibit valuable pharmacological properties. They can be used in analogy to known plasminogen activators in humans for the prevention or treatment of blood clots or other conditions where it is desirable to produce a local fibrinolytic or proteolytic activity via the mechanism of plasminogen activation, such as arteriosclerosis, myocardial and cerebral infarction, venous thrombosis, thromboembolism, post-operative thrombosis, thrombophlebitis and diabetic vasculopathy.

It was surprising to find that the novel hybrid PA proteins and mutants thereof according to this invention combine the beneficial properties of native t-PA and u-PA. Therefore, the new hybrid PA proteins and their mutants are fibrinolytically active. The unique fibrin-targeting properties, i.e. the possibility of activation of plasminogen, preferably in the presence of fibrin, is preserved. Furthermore, the new proteins have a prolonged in vivo stability compared to authentic t-PA.

The pharmaceutical preparations comprise a therapeutically effective amount of the active ingredient (hybrid PA or mutant thereof) together with organic or inorganic, solid or liquid pharmaceutically acceptable carriers suitable for parenteral administration, i.e. intramuscular, subcutaneous or intra-peritoneal, administration and which do not react in a harmful way with the active ingredients.

Suitable solutions for infusion, preferably aqueous solutions or suspensions, with the possibility of preparing these before use, for example from freeze-dried preparations containing the active ingredient alone or together with a carrier, such as mannitol, lactose, glucose, albumin and the like. The pharmaceutical preparations are sterilized and, if desired, mixed with additives, for example preservatives, stabilizers, emulsifiers, solubilizers, buffers and/or salts for regulating the osmotic pressure. Sterilization can be achieved by sterile filtration through the small pore size filter (0.45 pm diameter or less) and then the preparation can be freeze-dried, if desired. Antibiotics may also be added to preserve sterility.

The pharmaceutical preparations are distributed in unit dosage forms, for example ampoules, consisting of 1 to 2000 mg of a pharmaceutically acceptable carrier unit dose and about 1 to 200 mg, preferably about 5 to 100 mg, of the active ingredient per unit dose.

Depending on the type of disease and the age and condition of the patient, the daily dose given for the treatment of a patient weighing about 70 kg is in the range from 3 to 100 mg, preferably from 5 to 50 mg, per 24 hours. In the case of myocardial infarction, a dose of about 30 to 80 mg should preferably be given within 60 to 120 minutes, preferably in three portions and within about 90 minutes. The total amount of hybrid PA or mutant hybrid PA can also be given as pill injection.

The invention particularly includes hybrid vectors, transformed host strains and procedures for their production as described in the examples.

Brief description of the drawings

In the experimental section which follows, various embodiments of this invention are described with reference to the attached drawings in which: Fig. 1 and fig. 3 illustrates the nucleotide sequences and the deduced amino acid sequences of human t-PA cDNA and human u-PA cDNA, respectively. The first amino acids of the final proteins are underlined. Fig. 5 schematically illustrates the technique used to construct plasmid pEcoO.47AScaI. Fig. 6 schematically illustrates the construction of plasmid ph'tPAAScaI containing a mutated t-PA cDNA. Fig. 7 schematically illustrates the construction of plasmid pUNOtc which contains a cDNA insert consisting of the A-chain domains of u-PA and B-chain of t-PA. Fig. 8 schematically describes the construction of plasmid ptNC-UC which contains a cDNA insert consisting of the A-chain domains of t-PA and B-chain of u-PA. Fig. 9 schematically describes the construction of plasmid pDO2. Fig. 10 schematically illustrates the construction of plasmid pDO10 containing t-PA cDNA assembled with a beta globin fragment. Fig. 11 schematically illustrates the construction of plasmid pCGA26 containing t-PAc DNA under the control of the MCMV IE promoter and a beta globin fragment. Fig. 12 schematically illustrates the construction of t-PA expression plasmid pCGA28 and of universal expression plasmid pCGA44, both plasmids containing the neomycin resistance gene. Fig. 13 schematically illustrates the construction of t-PA expression plasmid pCGA42 and universal expression plasmid pCGA42d, both plasmids contain the hygromycin resistance gene. Fig. 14 schematically illustrates the construction of t-PA expression plasmid pCGA48 containing the neomycin resistance gene and the DHFR gene. Fig. 15 schematically illustrates the construction of expression plasmid pBR1a containing the mutated t-PA cDNA insert of plasmid ph'tPAAScaI. Fig. 16 schematically illustrates the construction of expression plasmid pBR2A containing a hybrid PA cDNA insert consisting of the A-chain domains of u-PA and B-chain of t-PA. Fig. 17 schematically describes the construction of u-PA expression plasmid pBR3a. Fig. 18 schematically illustrates the construction of expression plasmid pBR4a containing a hybrid PA cDNA insert consisting of the A-chain domains of t-PA and B-chain of u-PA. Fig. 19 schematically shows the construction of yeast expression vector pJDB207/ PH05-I-TPA containing PH05 promoter, invertase signal sequence and t-PA cDNA. Fig. 20 schematically illustrates the construction of plasmid pCS16. Fig. 21 schematically illustrates the construction of plasmid pCS16/UPA containing u-PA cDNA. Fig. 22 schematically shows the construction of plasmid PJDB207/ PH05-I-UPA. Figures 23-26 schematically illustrate the techniques used to convert primary hybrid PA constructs including the A-chain domains and the catalytic B-chain region of u-PA or t-PA into the finished constructs where the boundaries of the domains are at the active the site and/or at the natural exon-intron boundary sites: Fig. 23 shows the construction of a gene that codes for a hybrid PA consisting of the A-chain domains of t-PA and the B-chain of u-PA. Fig. 24 shows the construction of a gene which codes for a hybrid PA consisting of the A-chain domains of u-PA and the B-chain of t-PA. Fig. 25 shows the construction of a gene encoding a hybrid PA consisting of the u-PA growth factor domain, the kringle 2 domain of t-PA and the B chain of t-PA. Fig. 26 shows the construction of a gene which codes for a hybrid PA consisting of u-PA growth factor domain, kringle 2 domain of t-PA and B-chain of u-PA. Fig. 27 is a collection of hybrid PAs and mutant hybrid PAs exemplified in the experimental section.

Symbols used in the attached figures mean the following: AMP, Amp<R> ampicillin resistance gene (beta-lactamase) TET, Tet<R> tetracycline resistance gene

NEO Tn5 neomycin phosphortransferase

TN5PR bacterial promoter of transposon TN5

HPH hygromycin phosphotransferase

pBRori replication origin of plasmid pBR322

POIS 'poison sequence', pBR322 sequence that inhibits SV40 replication

SV40ori origin of replication to SV40, which falls

along with early and late promoters. SV40enh, SV40E 72 bp amplifier, part of SV40 early

promoter

HCMVE enhancer of human cytomegalovirus (HCMV)

main immediate early gen

MCMVP promoter/mRNA start site of mouse cytomegalovirus (MCMV) major immediate early gene

RSV Rous sarcoma virus LTR (promoter)

CAP position to 5' m7GP 'cap' to eukaryotic

mRNA

polyA polyadenylation site of mRNA

SPLD splice donor site, 5' end to intron

SPLA splice acceptor site, 3' end to intron BAP bacterial alkaline phosphatase

CIP calf intestinal phosphatase

(BamHl/Bgl2) Sau3a seat resulting from merger

of a BamHI and a Bglll site

Seal(part) mutated Seal seat

x < y restriction enzyme site x located with

the clock from y

p promoter

inv.SS invertase signal sequence

t transcription terminator

L links DNA

DHFR dihydrofolate reductase

mtPA Bowes melanoma t-PA

Experimental part,

Example 1: Introduction of a Seal site at the boundary between the kringle structures and the enzyme domain in human t-PA cDNA.

A method used to construct chimeras or hybrid molecules containing domains from both t-PA and u-PA consists of making desired restriction fragments derived from the respective clones, reassembling them in solution, then cloning the resulting constructs. After cloning, the structure of the chimeric molecules is verified by restriction mapping and DNA sequence analyses.

To obtain the hybrid molecules, both t-PA and u-PA cDNAs are cut at the boundaries between the respective kringle structures and the enzyme domains. This is achieved with u-PA by performing a partial cut with the restriction endonuclease Mstl, which separates the non-catalytic domain from the enzyme domain and associated sequences at its 3' end. No comparable suitable potential cutting seat is present in t-PA, and one is thus introduced as described below:

A) Construction of plasmid pEcoO. 47AScaI (see fig. 5)

In this construct, the unique Scal site (AGTACT) at nucleotide positions 940-945 of the t-PA cDNA is destroyed (AGTACT -♦ AGTATT) and another Scal site is introduced at nucleotide positions 963-968 (TCCACC -+ AGTACT ) at the 3' end of pretzel 2 (cf. figures 1 and 2). The coding of none of the amino acids is affected by these changes.

All restriction cuts are performed according to the manufacturers' (New England Biolabs, Bethesda Research Labs) instructions and the resulting cuts are analyzed by 3,556 polyacrylamide gel electrophoresis. The gel is stained with ethidium bromide (1.0 µg/ml) and visualized with ultraviolet light. The appropriate band is cut out and electroeluted in 0.5 x TBE (1 x TBE = 90 mM Tris-borate, pH 8.3, 2.5 mM EDTA). The electroeluted material is applied to an Elutip-d column (Schleicher and Schuell), the bound DNA is eluted in a high salt concentration and precipitated by adding ethanol. The precipitate is washed with ethanol, dried and dissolved in water.

Plasmid p¥349F (European Patent Application No. 143,081) containing human t-PA cDNA (synthesized from mRNA isolated from HeLaS3 cells and cloned into the Pstl site of plasmid pBR322) is cut with EcoRI and the 470 base-pair (bp) fragment (cf. Fig. 2) is isolated. The 150 bp EcoRI, Seal and the 290 bp EcoRI, Haelll fragments are obtained by cutting the 470 bp EcoRI fragment with Seal and Haelll, respectively. The two chains of the 470 bp EcoRI fragment are separated by denaturing the DNA in DMSO buffer (30% DMSO, 1 mM EDTA, 0.5$ xylene-cyanol, 0.05$ bromophenol blue) and electrophoresed on a 5% polyacrylamide gel in 0.5x TBE at 8 volts per centimeter [Maniatis et al., Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory; 1982]. The separated chains are recovered by electroelution followed by ethanol precipitation. A 31-mer deoxyoligonucleotide (incorporating the 5 desired nucleotide changes, cf. Fig. 5) is synthesized using the phosphotriester method. Fifty pmol of 31-mer is <32>P-labeled at the 5' end in a 20 µl reaction containing 1 x kinase buffer (10 x kinase buffer = 0.5 Tris-HCl, pH 7.5, 0.1 M MgCl2, 50 mM DTT, 1 mM spermidine, 1 mM EDTA), 30 pCi [α<32>P]ATP (Amersham, ~ 3000 Ci/mmol) and 10 units of T4 polynucleotide kinase (Bethesda Research Labs.). The reaction is incubated at 37°C for 30 minutes followed by the addition of 1 µl of 10 mM ATP, 10 units of T4 kinase and a further 30 minutes incubation at 37°C. The reaction is terminated by heating at 68°C for 10 min. The labeled 31-mer, with sequence identical to the non-transcribed chain, is used as a probe in a dot blot analysis [performed according to Zoller and Smith, Nucl. Acids Res., 10., 6487-6500

(1982); with the exception that the prehybridization and hybridization is done at 50°C and washing at 60°C] to determine which of the two chains hybridizes to it, i.e. represents the transcribed chain. The four DNAs are mixed together in a 20 µl fusion reaction consisting of 0.3 pmol of the transcribed strand, 2 pmol each of 150 bp EcoRI, Seal and 290 bp EcoRI, the Haelll fragments, 25 pmol of phosphorylated 31-mer and 1 x fusion buffer (5 x fusion buffer = 0.5 M NaCl, 32.5 mM Tris-HCl pH 7.5, 40 mM MgCl2 and 5 mM p<->mercaptoethanol). The mixture is incubated at 100°C for 3 min., 30°C for 30 min., 4°C for 30 min. and then on ice for 10 min. followed by the addition of 400 units of T4 DNA ligase (New England Biolabs) and the reaction incubated at 12.5°C overnight. The 470 bp fused fragment is recovered from a 3.55& polyacrylamide gel as described above and ligated to EcoRI cut and dephosphorylated pBR322 DNA (New England Biolabs) in 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 1 mM ATP, 1 mM spermidine, 0.1 mg/ml bovine serum albumin by overnight incubation at 12°C. The ligation mixture is used to transform competent E.coli strain HB101 (Maniatis et al., supra). Ampicillin-resistant colonies are selected on L-agar containing 50 µg/ml ampicillin and colonies containing the 470 bp fragment are identified by colony hybridization using the 31-mer as a probe [D. Woods, Focus 6, 1-3 (1984)]. Plasmid DNA is isolated on a small scale from several positive hybridization colonies [Holmes et al., Analyt. Biochem. 114, 193-197

(1981)] and the formation of the new Seal site is verified by combined EcoRI, Seal cutting. To ensure purity, plasmid DNA from the positive colonies is used a second time for transformation of E.coli HB101. Large scale plasmid production is made from such a second generation positive colony [Katz et al., J. Bacteriol. 114, 577-591 (1973); Biochemistry 16, 1677-1683 (1977)] and destruction of the original Seal site and formation of the new Scal site is verified by DNA sequence analysis using the method of Maxam and Gilbert [Methods Enzym. 65, 499-560 (1980)]. This plasmid is designated pEcoO.47AScaI. B) Reconstruction of human t-PA with mutant Seal site (see fig.

61

In this construct, the 470 bp EcoRI fragment present on wild-type human t-PA is replaced with the 470 bp EcoRI fragment containing the mutant Seal site. Plasmid pW349F containing human t-PA cDNA (see above) is cut with ClaI and the resulting sticky ends are blunted by the addition of 50 µm of dCTP, dGTP and 10 units of DNA polymerase I, Klenow fragment (Boehringer, Mannheim). The reaction is incubated at room temperature for 30 min. followed by phenol and ether extraction and ethanol precipitation. The precipitate is dissolved in water, cut with EcoRI and Seal, and 1.5 kg EcoRI, Seal and 4.3 kb Clal (blunt end), the EcoRI fragments are isolated. These two fragments are mixed with the 470 bp fragment from plasmid pEcoO.47AScaI after EcoRI cutting and ligated as described above at 12°C overnight. Competent E.coli HB101 cells are transformed with ligation mixture and tetracycline-resistant colonies selected on L-agar containing 12.5 pg/ml tetracycline. The colonies containing the 470 bp mutant fragment are identified by colony hybridization using the aforementioned 31-mer as probe. DNA from the minilysates of several of the positive hybridization colonies is prepared and the exact nature of the construct is verified by performing appropriate restriction cuts. Such a plasmid with the desired changes is called ph-PAAScal.

Example 2: Construction of a u-PA/t-PA hybrid molecule; plasmid pUNC*tc (see Fig. 7)

This construct is a hybrid between the non-catalytic region of u-PA (containing the 5' non-coding region, signal, growth factor and coil sequences) and the catalytic or enzyme domain of human t-PA.

Urokinase cDNA is prepared from mRNA from human Eep3 cells [cf. T. Maniatis et al., Molecular Cloning (1982), pp. 188-246]. A 1.3 kb SmaI-BamHI fragment and a 1 kb BamHI-EcoRI fragment of u-PA cDNA are cloned into the SmaI, EcoRI sites of pTJN121 [B. Nilsson et al., Nucl. Acids Res. 11, 8019-8030

(1983)] leads to plasmid pcUK176. The restriction endonuclease map to the human u-PA cDNA insert is shown in fig. 4. The nucleotide sequence and deduced amino acid sequence of the u-PA insert is given in fig. 3.

Plasmid pcUK176 is cut with Xmal (cf. Fig. 4; Xmal is an isoschizomer of Smal) and Mstl and the 521 bp fragment is isolated. Restriction enzyme Mstl detects the DNA sequence TGC^GCA (arrows indicate the site of the cut) and forms blunt ends after the cut; this enzyme therefore cuts out the u-PA cDNA at nucleotides 520-525, i.e. immediately after the last cysteine residue (amino acid 131) containing the kringle [Holmes et al., Biotechnology 3, 923-929 (1985)], thereby separating the sequence coding for non-catalytic and catalytic regions.

Plasmid ph-tPAAScal is cut with Seal and Hind III (Hind III is present in the vector) and the 1.8 kb fragment is recovered. Restriction enzyme Seal recognizes the DNA sequence TGC^GCA (arrows indicate the site of the cut) and also produces blunt ends after the cut. Seal cuts pbvtPAAScal DNA after serine residue 262 [1 amino acid after the last cysteine in pretzel 2; Pennica et al., Nature 301, 214-221 (1983)], thereby separating the non-catalytic and catalytic domains.

The two fragments are mixed and ligated to XmaI, HindIII cut pUC18 vector DNA. After transformation of E.coli HB101, the colonies having the correct insert are identified by colony hybridization using the 2.0 kb Bglll fragment of human tPA (cf. Fig. 2) as probe [the probe is labeled using the "random-starter method ": Feinberg et al., Analyt. Biochem. 132, 6-13 (1983)]. The DNA sequence at the border of the liger neither of the u-PA and t-PA fragments is verified by DNA sequence analysis. A correct clone is designated pUNC'tc.

Example 3: Construction of a t-PA/u-PA hybrid molecule: plasmid ptNC-UC (cf. fig. 8)

This construct is just the opposite of pUNC*tc where the non-catalytic region of ph'tPAAScaI (containing the 5' non-coding region, leader, finger, growth factor, kringle 1 and kringle 2 domains) is fused to the catalytic domain of human u -ON. Plasmid ph*tPAAScaI is cut with SacI and Seal (cf. Fig. 8) and an approximately 1.0 kb fragment is isolated. Plasmid pcUK176 is first cut with BamHI and then partially cut with Mstl and the approximately 800 bp fragment isolated. Afterwards, the BamHI cut sequence is cut with EcoRI and the approximately 1.0 kb fragment isolated. These three fragments are mixed with pUC19 vector cut with SacI, EcoRI and ligated. E.coli HB101 is transformed with the ligation mixture and colonies having the correct insert are identified by colony hybridization using the same 2.0 kb BglII probe described above. The DNA sequence at the boundary between t-PA and u-PA DNA has been verified by DNA sequence analysis. A proper clone is named ptNC-UC.

Example 4: Construction of an express. lon vector for use in mammalian cells.

A. Conversion of the HgiAI site in the t-PA cDNA into a HindIII site This is done in five steps (Fig. 9).

Plasmid p¥349F (European Patent Application No. 143,081) is partially cut with the restriction enzyme HgiAI by incubating 20 µg/ml DNA for 1 hour at 37°C with 12 U/ml enzyme in the buffer recommended by the manufacturer (Bethesda Research Laboratories). except that it is supplemented with 10 pg/ml ethidium bromide to suppress secondary cutting of the plasmid. Linearized plasmid DNA is then applied to a 0.8% agarose gel in TBE buffer (TBE: 89 mM Tris-borate pH 8.9 as containing 1 mM EDTA), eluted electrophoretically in the same buffer, extracted twice with phenol, twice with chloroform and finally precipitated with alcohol at -20°C after addition of 0.1 vol. of 3 M sodium acetate pH 5.2. DNA in the precipitate is dissolved with 0.2 mg/ml in TE (TE: 10 mM Tris-HCl pH 7.2 with 0.1 mM EDTA).

63 µl of linearized DNA is then incubated for 30 min. at 37°C with 15 U T4 DNA polymerase in ligase buffer [33 mM Tris-acetate (pH 7.9), 66 mM potassium acetate, 10 mM magnesium acetate, 0.5 mM dithiothreitol and 0.1 mg/ml bovine serum albumin] followed by heating for 10 min. at 60° C to inactivate the enzyme. The purpose of this incubation is to use the exonucleolytic activity of T4 polymerase to remove the four nucleotides that stick out after cutting with HgiAI to obtain blunt-ended DNA molecules.

To ligate the HindIII linkers (CAAGCTTG) to blunt-ended DNA 6 µl (300 ng) kinase-treated linkers are added to the above solution with 4 µl 10 mM ATP and 3 µl t4 DNA ligase (New England Biolabs, 400 u/µl) followed by of 16 h incubation at 16°C. The ligation is terminated by heating the mixture for 10 min. at 68° C, then the DNA is cut with HindIII and BglII, i.e. 15 µl (135 U) HindIII is added with 1.5 µl 4M NaCl, 0.2 µl IM MgCl2 and 11 µl of 1 mg/ml bovine serum albumin, incubated at 37°C in oxygen followed by addition of 40 U of BglII followed by 1 h incubation at 37°C. The resulting 177 base pair fragment is purified on a b% polyacrylamide gel run in TBE, eluted in TNE (TNE: 10 mM Tris-Hcl pH 8.8 containing 100 mM NaCl and 1 mM EDTA), absorbed to DEAE cellulose (Whatman DE52), eluted with 1 M NaCl in TNE, diluted with 4 volumes of water, precipitated at -20°C after addition of 2.5 volumes of ethanol and then dissolved in 17 µl TE (TE: 10 mM Tris-HCl pH 8.0 containing 1 mm

EDTA).

Plasmid pRSVneo is derived from plasmid pSV2neo [P.J. Southern and P. Berg, J. Mol. Appl. The gene. 1, 327-341 (1982)] where the SV40-derived PvuII-HindIII fragment has been replaced with a PvuII-HindIII fragment containing the Rous sarcoma virus LTR promoter in the same way that pRSVcat was constructed from pSV2cat [CM. Gorman et al., Proe. Nati. Acad. Pollock. USA 79, 6777-6781 (1982)]. 5 µg of this plasmid is cut in a 50 µl volume with 24 U of BglII according to the manufacturer's instructions. After 1 h of incubation at 37° C, 40 U HindIII is added and the incubation is continued for 1.5 h, after which a 5.4 kb fragment is purified as described above.

17 µl of the purified 177 bp fragment is ligated for 18 hours at 16°C to 2 µl (20 ng) of pRSVneo fragment with 0.25 µl (100

U) T4 ligase in a total volume of 22 µl of ligase buffer, after which plasmid DNA is used to transform E.coli according to D. Hanahan [J. Mol. Biol. 166, 557-589 (1983)]. From the resulting ampicillin-resistant strains, one is selected containing the plasmid designated ptPAL with the 177 bp HindIII-BglII fragment, as evidenced by restriction analysis. 0.1 pg of this plasmid is cut in 60 µl with 16 U of BglII as recommended by the manufacturer for 1.5 h at 37° C. To this solution is then added 20 U of calf intestinal alkaline phosphatase (Boehringer Mannheim) and the incubation continues for 30 min. after which the DNA is extracted twice with phenol, twice with chloroform and precipitated after adding 0.1 volume of 3.0 M sodium acetate pH 5.2 and 0.6 volume of isopropanol, dissolved in TE, further purified by agarose gel electrophoresis as described above, extracted twice with phenol, twice with chloroform, precipitated at -20°C after addition of 2.5 volumes of ethanol and 0.1 vol. 3 M sodium acetate pH 5.2 and then dissolved in 30 µl TE. The 2.1 kb tPA Bglll fragment is then excised from 5 pg pW349F in a 25 µl reaction mixture using 20 U BG1II for 2 h at 37°C, purified on a 0.8$ agarose gel, electrophoretically eluted as described above, extracted twice with phenol, twice with chloroform, precipitated at -20°C after addition of 2.5 volumes of ethanol and 0.1 vol. 3 M sodium acetate pH 5.2 and dissolved at a concentration of 8 ng/pl in TE. 1 µl of the t-PA fragment is then ligated in a 10 µl reaction to 7.5 ng BglII cut vector DNA with 100 U T4 ligase (Biolabs) for 17 h at 16°C and then transformed into E.coli. One of the resulting clones, designated pDO2, contains the t-PA BglII fragment inserted in such a way that the plasmid contains a continuous open reading frame for human t-PA. B) Combining t-PA cDNA with the beta-globin fragment

Plasmid pDO10 (Fig. 10) is constructed by assembly of three DNA fragments: (i) a 2.1 kb fragment starting with a HindIII site and terminating with a BglIII site containing the entire t-PA coding sequence is isolated from an agarose gel which is added 10 pg of pDO2 DNA cut partially with BglII and completely with HindIII. (ii) pUB is a plasmid containing rabbit beta-globin gene [A. Van Ooyen et al., Science 206, 337

(1979)] subcloned as a BglII partially cut fragment into the BamHI site of plasmid pUC9 [J. Vieira and J. Messing, Gene 19, 259-268 (1982); ibid. 19, 269-276 (1982)]. From this plasmid, a 1.2 kb BamHI-HindlII fragment containing the second intron and the polyadenylation site is excised and purified by agarose gel electrophoresis. (iii) Vector pDO1 is constructed, in counter-clockwise order from the HindIII site (Fig. 10) to the HindIII-AccI fragment of pBR322 containing the origin of replication, a 0.3 kb fragment containing the enhancer of human cytomegalovirus (HCMV) terminating in a synthetic XbaI site followed by one more copy of this enhancer attached to the homologous promoter terminating at a synthetic HindIII site. This vector DNA is cut with HindIII and the 6.3 kb linear plasmid is purified by agarose gel electrophoresis.

C) Insertion of the tPA/globin combination into pSP62Pst33

(see figure 11)

pSP62Pst33 (Fig. 11) is a plasmid containing a 2.1 kb Pstl fragment of mouse cytomegalovirus (MCMV) DNA, containing the viral immediate early (IE) promoter, inserted into the Pstl site of plasmid pSP62 (Boehringer Mannheim) as indicated in the figure . Into the HindIII site of pSP62Pst33, the HindIII fragment from pDO10 is inserted. A plasmid, pCGA26, has been selected where the t-PA coding sequence has been inserted so that it can be transcribed in the correct orientation from the MCMV IE promoter.

Dl Insertion of the MCMV/tPA/globin unit into pFASV2911neo

(see figure 12)

Plasmid pSV2911neo [F. Asselbergs et al., J. Mol. Biol. 189, 401-411 (1986)] which contains the neomycin (neo) phosphotransferase gene from transposon TN5 in an SV40 expression cassette (Fig. 12). It thus exerts resistance to neomycin and kanamycin when introduced into mammalian tissue culture cells. pSV2911neo DNA is prepared for cloning by cutting with BamHI, treated with calf intestinal alkaline phosphatase, two extractions with phenol, two with chloroform, precipitation with alcohol and finally dissolved in TE. Plasmid pCGA26 is cut with restriction enzyme AccI, which cuts the sequence GT/ACAC at position 345 in the MCVM enhancer/promoter region [K. Doersch-Haessler et al., Proe. Nati. Acad. Pollock. USA 82, 8325-8329 (1985)] and the sequence GT/CGAC (can also be cut with Sali) after the globin portion. The two base overhangs that result after cutting are filled in with E.coli (large fragment) DNA polymerase I, the now blunt ends are ligated to BamHI linkers (CGGATCCG) and these are cut with the BamHI enzyme. The 3.8 kb fragment carrying the MCMV/tPA/globin unit now with BamHI ends is purified via an agarose gel and then ligated to pSV2911neo DNA prepared as described above leading to expression plasmid pCGA28.

E) Expression vectors derived from pCGA28

pCGA42 is a derivative of pCGA28 where the neo coding sequence (between the Bglll site and the SmaI site) is replaced by the coding sequence of the hygromycin resistance gene. This has been achieved (see Fig. 13) by cutting plasmid pSV2911neo at its unique Smal site, and ligation of a Bglll linker (CAGATCTG) to DNA followed by cutting with Bglll. The resulting large DNA fragment consisting of the vector minus the neocoding sequence is purified on an agarose gel and ligated to the small BamHI fragment from plasmid pLG89 [L. Gritz et al., Gene 25, 179-188 (1983)] also purified on an agarose gel, which

leading to plasmids pCGA25c and pCGA25d, which contain the hygromycin phosphotransferase gene in the correct and non-correct orientations, respectively. When transfected into CHO DUKXB1 cells under standard conditions (see Example 16), pCGA25C yields 60 colonies/µg DNA resistant to 0.2 µg/ml hygromycin B, a concentration that kills CEO cells containing a plasmid that does not encode hygromycin -resistance, for example pCGA28. In pCGA25c, the sequences that code for hygromycin-B resistance are located so that in E.coli they are transcribed from the Tn5 promoter (which in transposon Tn5 transcribes the kanamycin resistance gene). A 2.5 ml culture of "Luria broth" (LB) containing 40 mg/l hygromycin-B inoculated with 0.05 ml of an overnight, i.e. saturated, culture of E.coli DH1 bacteria (grown under 50 mg/l ampicillin selection) reaches after 3 h aerated culture at 37°C a bacterial density at least 10 times higher than that reached by the bacterium with plasmids that do not contain a hygromycin gene functional in E.coli, such as pCGA25d, pCGA28 or pAT153 [A.J. Twigg et al., Nature 283, 216-218 (1980)3, is tested. The functionality of the hygromycin-B resistance gene both in animal tissue culture cells and in E.coli greatly facilitates the use of plasmid pCGA25c and its derivatives. Plasmid pCGA42 is then constructed by inserting the BamHI fragment from pCGA28 containing the MCMV/t-PA/beta-globin cassette into pCGA25C. This transfers the gene that expresses t-PA into cells that cannot be transformed into geneticin resistance or that are already geneticin resistant. pGA42 can also express the hygromycin gene in E.coli, this means that pCGA42 containing E.coli DH1 can grow to densities at least 10 times higher than, for example, pCGA28 containing E.coli when tested as it is described above.

Plasmid pCGA28 contains two SacI sites, one early part of a linker just behind the MCMV promoter, the other in the t-PA cDNA. The sequence between the SacI sites has been removed by cutting first with the restriction enzyme, purification of the larger fragment via an agarose gel and circularization of this linear DNA using T4 DNA ligase, which forms plasmid pCGA44 (see fig. 12). Any cDNA cloned in the correct orientation into the now unique SacI site of pCGA44 effectively replaces the t-Pa coding sequence of pCGA28 and is efficiently expressed.

pCGA42d is derived from pCGA42 by removing the 1.4 kb SacI fragment (see Fig. 13). Into the now unique SacI site, cDNAs different from the t-PA cDNA can be inserted and expressed at a high level in tissue culture cells.

Example 5: Insertion of u-PA. t- PA and hybrid PA cDNA into express. ion vector pCGA28

A) Insertion of t-PA cDNA (see figure 15)

In this construct, the t-PA cDNA fragment from plasmid ph-tPAAScal is inserted into pCGA28. This construction is considered necessary to act as a control for changes that may have occurred during the restructuring of the Seal seat. The 1.4 kb SacI fragment is isolated from plasmid ph-tPAAScal after SacI cutting. Expression vector pCGA28 is also cut with SacI and the 8.2 kb vector fragment is isolated and dephosphorylated in 100 µl reaction mixture containing 0.1 mM Tris pH 8.0, 0.1% SDS and 0.02 units of bacterial alkaline phosphatase. After incubation at 60°C for 30 min, the reaction is extracted twice with phenol and ether and then precipitated with ethanol. The precipitate is dissolved in water and part of it is used for fusion to the 1.4 kb SacI fragment from ph-tPAAScal. The ligation mixture is used to transform E.coli EB101 and minilysate DNA prepared from ampicillin-resistant colonies is cut with appropriate restriction enzymes to verify that the SacI inserted fragment is in the desired orientation. The plasmid that has the desired orientation is called pBRlA. The plasmid with the SacI fragment in the opposite orientation is called pBRlB.

B) Insertion of hybrid UPA^tpaS cDNA (see figure 16).

In this construct, the hybrid UPA<A>TPA<B> cDNA fragment from plasmid pUNC.tc is inserted into expression vector pCGA28. The pUNC.tc DNA is cut with Smal (cf. Figure 7), and the 1.24 kb fragment is isolated and ligated to the SacI cut, dephosphorylated 8.2 kb pCGA28 vector DNA. The E.coli HB101 cells are transformed with the ligation mixture and colonies containing the SacI insert in the desired orientation identified by performing restriction cuts on minilysate DNA. Plasmid with pUNC.tc DNA inserted in the desired orientation is designated pBR2A and the one with the opposite orientation pBR2B.

C) Insertion of u-PA cDNA (see Figure 17)

In this construct, human u-PA DNA is inserted into expression vector pCGA28 and together with pBR1 this plasmid serves as a parental plasmid control and confirms the utility of pCGA28-type vectors. Plasmid pcUK176 is cut with Smal, AhalII (cf. Fig. 4), the 2.25 kb fragment is isolated, and ligated to the phosphorylated SacI linker as described above. After the SacI cut, the 2.25 kb fragment is isolated and ligated to the SacI cut, dephosphorylated 8.2 kb pCGA28 DNA fragment. E.coli HB101 is transformed and colonies containing desired plasmid identified by cutting minilysate DNA with restriction enzymes. The plasmid with human u-PA DNA in the correct orientation is called pBR23A and that in the opposite orientation pBR3B.

D) Insertion of hybrid TPA^ UPA^ cDNA ( Fig. 18 )

Here, hybrid TPA<A>UPA<B> cDNA from plasmid ptNC.UC is inserted into expression vector pCGA28. The 2.75 kb Smal (present in the vector), AhalII fragment is isolated from ptNC.UC DNA, ligated to phosphorylated SacI linker, the linker 2.75 kb fragment is isolated and ligated to SacI cut, dephosphorylated vector DNA and the desired colonies identified as described above . The plasmid with the ptNC.UC DNA insert in the correct orientation is called pBR4A.

Example 6: Construction of a g. year express. ion vector containing the PH05 promoter. invertase signal sequence and t-PA coding region

A) Synthesis of oligodeoxyribonucleotides for invertase signal sequences

Four oligodeoxyribonucleotides: I-1, 1-2, 1-3, 1-4 are synthesized using a DNA synthesizer (model 380B Applied Biosystems). After deblocking, the synthetic fragments are purified on a 12% polyacrylamide gel containing 8 M urea. Salt-free pure oligodeoxyribonucleotides are obtained using Sep. Pak (Waters Associates). These fragments consist of duplexes encoding the invertase signal sequence with commonly used yeast codons.

B) Subcloning of the invertase signal sequence in plasmid p31

a) Preparation of vector:

1.5 pg of p31R/SS-TPAA2 (Europe patent application no. 143,081)

is cut with 10 U EcoRI (Boehringer) in 50 µl of 10 mM Tris.HCl pH 7.5, 6 mM MgCl2, 100 mM NaCl, 6 mM mercaptoethanol for one hour at 37"C. After addition of 1 µl of 2.5 M NaCl, 10 U Xhol (Boehringer) is added and incubated at 37°C for a

hour. The 4.2 kb vector is isolated on a 0.8% preparative agarose gel. The gel piece is transferred to a Micro Colloidor tube (Sartorius GbmE), covered with 200 µl TE and electroeluted (electrophoresis at 90 mA for 50 min.). The TE solution is collected and precipitated in 2.5 volumes of absolute ethanol after the addition of 0.1 volumes of 10 x TNE. The DNA precipitate is washed with cold 80% ethanol and dried in vacuum. DNA is resuspended in 6 µl TE (40 pmol/µl). b) Fusion of oligodeoxyribonucleotides (I-l. 1-2. I-3. 1-4). treatment with kinase and ligation with vector.

A solution containing 10 pmol of each of the four deoxyribonucleotides in 10 µl of 0.5 M Tris.HCl pE 8 is incubated at 95°C for 5 minutes in a water bath. The water bath is cooled slowly to 30° C over a period of 5 hours. To this fusion mixture is added 2 µl each of 0.1 M MgCl 2 , 0.1 M NaCl, 30 mM DTT, 4 mM ATP and 8 TJ (1 µl) polynucleotide kinase (Boehringer). The kinase treatment is carried out at 37°C for one hour. The pooled kinase-treated oligodeoxyribonucleotides and 60 pmol of p31R/SS-TPAA2 cut vector (1.5 µl) are ligated with 400 U (1 µl) T4 DNA ligase (Biolabs) at 14°C for 17 hours. The reaction is stopped by incubation at 65°C for 10 min. 10 µl of this ligation mixture is used for transformation of E.coli HB101 Ca<++> cells [M. Dagert and S.D. Ehrlich, Gene 56, 23-28

(1979)]. 20 amp<R> colonies are picked. DNA is prepared using the rapid isolation procedure (D.S. Eolmes and M. Quigley. Anal. Biochem. 114, 193-197 (1981)]. DNA is cut with EcoRI and Xhol, radiolabeled with EcoRI end and analyzed on b% polyacrylamide gel containing 8 M urea using radiolabeled pBR322 Eaelll cut DNA as a marker. Bands of the correct size are observed from DNA obtained from all 20 clones. One clone was grown in 100 ml LB medium containing 100 pg/ml ampicillin. Plasmid DNA is isolated and is referred to as p31RIT-12.

C) Construction of PJDB207/PH05-I-TPA (see fig. 19)

a) Preparation of vector:

Three pg pJDB207/PH05-TPA18 (Europe patent application no. 143,081)

is incubated at 37°C for one hour with 10 U BamHI in 50 µl 10 mM Tris.HCl pH 7.5, 6 mM MgCl2, 100 mM NaCl, 6 mM mercaptoethanol. A section is checked on a 1$ agarose gel in TBE buffer to determine complete cutting. The cutting mixture is incubated at 65°C for 10 min. Then 0.5 µl 5 M NaCl is added followed by 15 U Xhol (Boehringer). This is incubated at 37°C for one hour. 6.8 kb vector is isolated on a 0.8$ preparative agarose gel. DNA is extracted by electroelution and after precipitation dissolved in TE.

b) Xhol cutting of P31/ PH05-TPA18:

Thirty pg p31/PH05-TPA18 (Europe patent application no. 143,081)

was incubated at 37°C for one hour with 60 U Xhol (15 U/µl) in 200 µl 10 mM Tris.HCl pH 8, 6 mM MgCl2, 150 mM NaCl, 6 mM mercaptoethanol, extracted with an equal volume of phenol-chloroform , and precipitated in ethanol.

c) Partial Pstl cutting of Xhol cut p31/ PH05- TPA18

The precipitated Xhol cut p31/PH05-TPA18 DNA is resuspended in 250 µl 10 mM Tris-HCl pH 7.5, 6 mM MgCl2, 50 mM NaCl, 6 mM mercaptoethanol, 2.5 mg ethidium bromide, incubated at 37°C for 35 minutes with 22.5 U Pstl, and extracted with an equal volume of phenol, followed by an equal volume of chloroform-isoamyl alcohol (50:1). The 1.6 kb fragment is isolated on a 1% preparative agarose gel. DNA is extracted by electroelution and precipitated [Inset 1].

d) Sall-Xhol cleavage of P31RIT-12:

Thirty pg of p31RIT-12 is incubated at 37°C for one hour with 60 U

Sali (Boehringer 12 U/pl) and 60 U Xhol (15 U/pl) in 200 pl 10 mM Tris.HCl pH 8, 6 mM MgCl2, 150 mM NaCl, 6 mM mercaptoethanol, extracted with an equal volume of phenol-chloroform and precipitated in ethanol. The 869 bp fragment is isolated on a 1.2$ preparative agarose gel. DNA was extracted by electroelution, desalted over DE-52, and precipitated in ethanol.

e) Hgal cutting of Sall-Xhol cut P31RIT-12

SalI-XhoI cut p31RIT-12 is resuspended in 100 µl 6 mM

Tris.HCl pH 7.5, 10 mM MgCl2, 50 mM NaCl, 1 mM dithiothreitol, 10 mg bovine serum albumin and is incubated at 37°C for one hour with 6 U Hgal (Biolabs, 0.5 U/pl). The 600 bp fragment is isolated on a 1.2% agarose gel. DNA is extracted by electroelution and precipitated in ethanol.

f) Base pairing of linker oligonucleotides

90 pmol of two oligodeoxyribonucleotides with the following

sequences

Hgal Pstl

5' CTGCATCTTACCAAGTGATCTGCA 3'

3'AGAATGGTTCACTAG 5'

is suspended in 10 µl of 0.5 mM Tris.HCl pH 8 in a siliconized Eppendorf tube. The solution is incubated at 95°C for 5 min. and then slowly cooled to room temperature overnight.

g) Kinase treatment of linkers

To the above solution is added 2 pl 0.1 M KC1, 2 pl

0.1 M MgCl2, 3 µl 30 mM DTT, 1 µl 200 mM ATP, 8 U polynucleotide (8U/µl). This is incubated at 37°C for one hour. h) Ligation of the Hgal fragment from p31RIT-12 with the kinase-treated linkers.

The kinase-treated linker solution is transferred to a tube containing the dry Hgal fragment, and 400 U of T4 DNA ligase is then added. The solution is incubated at room temperature (21-22°C) for 90 minutes, diluted to 100 µl with TE and extracted with an equal volume of phenol-chloroform. The fragment is precipitated by adding 0.6 volume of isopropanol and 0.1 volume of 3 M sodium acetate at room temperature to the aqueous solution.

i) BamHI-PstI cutting of the above solution.

The above dried DNA is cut with 10 U BamHI and 10 U Pst I in 20 µl 10 mM Tris.HCl pH 7.5, 100 mM MgCl 2 , 6 mM mercaptoethanol for one hour at 37°C. After dilution to 100 µl, the solution is extracted with an equal volume of phenol-chloroform, and the aqueous layer is precipitated in isopropanol [inset 2].

j) Ligation of the three fragments

100 fmol of pJDB207/PH05-TPA18 BamHI-XhoI cut vector fragment, 200 fmol of each of the other two insert fragments [1 and 2] are ligated in 10 µl of 50 mM Tris.HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400 U T4 DNA ligase for 16 hours at 15°C. The reaction is stopped by incubation at 65°C for 10 min. 5 µl of this alloy mixture is used for transformation of E.coli HB101 Ca<++> cells. 10 amp<R> colonies were picked and DNA prepared using the rapid isolation procedure. After analysis with EcoRI, PstI and BamHI-HindIII, fragments of the correct size were observed. A clone was grown in 100 ml LB medium containing 100 µg/ml ampicillin. Plasmid DNA is isolated and is referred to as pJDB 207/PH05-I-TPA.

Example 7: Construction of plasmid pCS16/UPA which contains the region which codes for u-PA.

A) Construction of plasmid pCS16 (see Figure 201.

A 1.5 kb Pstl-BamHI fragment of plasmid pUN121 [B. Nilsson et al., Nucl. Acids Res. 11, 8019-8030 (1983)] containing the cl gene of phage lambda and parts of the tetracycline resistance gene have been cloned into pUC18 [J. Norrander et al., Gene 26, 101-106 (1983)], and cut with Pstl and BamHI. The resulting clone is cut with Pstl. The 3' overhangs are removed in a reaction with T4 DNA polymerase and the Xhol linkers are ligated to the blunt ends. After cutting with Xhol, the molecule is recycled by ligation. A sample of the alloy mixture is used to transform Ca<++> treated E.coli HB101 cells. The DNA of individual ampicillin-resistant, transformed colonies was analyzed. One of several correct clones is selected and referred to as pCS16.

B) Construction of plasmid pCS16/ UPA (see Figure 21)

Urokinase cDNA as it exists in plasmid pcUK176 (see example 2) is subcloned into plasmid pCS16. The subcloned cDNA extends from the SmaI site in the 5' untranslated region (Fig. 4) to the PvuII site at nucleotide positions 1439-1444 in the 3' untranslated region (with numbering according to Fig. 3). 15 pg of plasmid pcUK176 is cut with PvuII. The 379 bp PvulI fragment is isolated from other fragments on a 1.5% agarose gel in Tris-borate-EDTA buffer pH 8.3. DNA is electroeluted, purified on DE52 (Whatman) ion-exchange chromatography and precipitated with ethanol. 1.2 pg of single chain Xhol linkers (5'-CCTCGAGG-3') are phosphorylated at the 5' ends, heated for 10 min. at 75°C, fused upon cooling to room temperature and stored at -20°C. 0.9 pg of kinase-treated, double-stranded Xhol linkers are ligated at an 80-x molar excess to the blunt ends of the 379 bp PvuII fragment of pcUK176 (see above) in 20 µl of 60 mM Tris.HCl pH 7.5, 10 mM MgCl2. 5 mM DTT, 3.5 mM ATP and 400 units of T4 DNA ligase (Biolabs) at 15°C for 16 hours. The mixture is heated for 10 min. at 85°C. Excess linker molecules are removed by precipitation with 0.54 volumes of isopropanol in the presence of 10 mM EDTA and 300 mM sodium acetate pH 6.0 for 30 min. at room temperature. DNA is cut with XhoI and BamHI. A 121 bp BamHI-XhoI fragment is isolated on a 1.5% agarose gel in Tris-borate-EDTA buffer pH 8.3. 6 pg of plasmid pcUK176 was cut with SmaI and BamHI. A 1.3 kb Smal-BamHI fragment containing most of the u-PA coding sequence is isolated. 6 µg of plasmid pCS16 was cut with SmaI and XhoI. The 2.7 kb vector fragment was isolated. The DNA fragments are electroeluted from the gel and ethanol precipitated. 0.2 pmol of the 1.3 kb Smal-BamHI fragment, 0.2 pmol of the 121 bp BamHI-Xhol fragment (both fragments together consist of the entire u-PA coding sequence) and 0.1 pmol of the 2.7 kb vector fragment are ligated in 10 µl 60 mM Tris.HCl pH 7.5 , 10 mM MgClg, 5 mM DTT, 3.5 mM ATP and 400 units of T4 DNA ligase at 15°C. One and 3 µl samples of the incubation mixture are added to 100 µl of Ca<++ >treated E.coli HB101 cells. The transformation is performed as described [A. The membrane et. al., Proe. Nati. Acad. Pollock. US 75, 1929 (1978)]. 12 ampicillin-resistant colonies are grown in LB medium containing 100 mg/l ampicillin. DNA is isolated according to Holmes et al., [Anal. Biochem. 114, 193 (1981)] and analyzed by EcoRI, PvuII and XhoI restriction digestion. A clone with the expected restriction fragments is referred to as pCS16/UPA.

Example 8: Construction of plasmid pJDB207/PHO5-I-UPA

(fig. 22)

pJDB2 0 7/ PHO 5-1-UPA contains the PH05 promoter, the invertase signal sequence, the coding sequence for finished urokinase and

PH05 transcription terminator in a tandem array, cloned into the pJDB207 yeast expression vector. 20 pg of plasmid pCS16/UPA is cut completely with 40 units of EcoRI. After phenol extraction and ethanol precipitation, the EcoRI cut DNA is further cut with Taql at 65°C. The resulting fragments are separated on a preparative 1.2% agarose gel. The 462 bp TaqI-EcoRI fragment is isolated by gel electroelution and ethanol precipitation.

An oligodeoxyribonucleotide linker of the following formula

(I) 5'-CTGCAAGCAATGAACTTCATCAAGTTCCAT-3 *

(II) 3'- TCGTTACTTGAAGTAGTTCAAGGTAGC-5'

is ligated to the Taql site on the DNA fragment. The linker restores the 5' terminal end of the coding sequence of finished u-PA (nucleotides 130-154, Fig. 3) and establishes the correct reading frame at the fusion to the invertase signal sequence. The 5'-CTGCA sequence of the linker fills the corresponding 3' truncated end of the invertase signal sequence formed by Hgal cutting.

300 pmol of each of the ol igodeoxynucleotides I and II are phosphorylated and joined. 5.25 pg (600 pmol) of phosphorylated, double-stranded linker DNA is ligated to 1.7 pg (5.6 pmol) of the 462 bp TaqI-EcoRI fragment (see above) in 175 µl 60 mM Tris.HCl pH 7.5, 10 mM MgCl2, 1 mM ATP , 5 mM DTT and 800 units of T4 DNA ligase at 15 °C for 16 h. T4 DNA ligase is inactivated for 10 min. at 85° C. Linkers that are in excess are removed by precipitation in the presence of 10 mM EDTA, 300 mM sodium acetate pH 6.0 and 0.54 volume isopropanol. DNA is cut with Pstl. A unique 312 bp fragment containing the linker linked to the DNA sequences encoding u-PA up to nucleotide 436 (Pstl site, see Fig. 3) is isolated. The DNA fragment is purified by electroelution and precipitation with ethanol.

Plasmid pCS16/UPA is cut with XhoI and PstI. A 1007 bp Pstl-Xhol fragment is isolated and purified. This fragment contains most of the coding sequence for urokinase.

Plasmid p31RIT-12 (see Example 6B) is cut with SalI and XhoI. An 882 bp SalI-XhoI fragment is isolated from the gel by electroelution and ethanol precipitation. The fragment is further cut with BamHI and Hgal. A 591 BamHI-Hgal fragment containing the PH05 promoter region and the invertase signal sequence is isolated.

Plasmid pJDB207/PH05-TPA18 (see European Patent Application No. 143,081) be cut with BamHI and XhoI. The 6.8 kb vector fragment is isolated on a preparative 0.6% agarose gel in Tris-acetate buffer pH 8.2. DNA is electroeluted and precipitated with ethanol.

All the DNA fragments are resuspended in H 2 O at a concentration of 0.1 pmol/µl. 0.2 pmol of 591 bp BamHI-Hgal fragment, 0.2 pmol 312 bp Hgal-Pstl fragment, 0.2 pmol 1007 bp Pstl-Xhol fragment and 0.1 pmol 6.8 kb BamHI-XhoI vector fragment were ligated for 15 h at 15°C at 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 5 mM DTT, 1 mM ATP and 400 units of T4 DNA ligase. One µl of the ligation mixture is used to transform E.coli HB101 Ca<++> cells. 12 amp<R> colonies are picked and grown in LB medium containing 100 mg/l ampicillin. DNA is prepared using the rapid isolation procedure [D.S. Holmes et al., Anal. Biochem. 114, 193 (1981)]. Upon restriction cutting of plasmid DNA with HindIII and EcoRI, the expected restriction fragments were observed. Plasmid DNA from a single clone was selected and referred to as pJDB207/PH05-1-UPA.

Example 9: A t-PA/u-PA hybrid plasminogen activator with the t-PA A-k.ide domains and u-PA B-k.ide (primary DNA construction)

Another method for the construction of an in-frame fusion of DNA sequences encoding the A-chain of t-PA and the B-chain of u-PA at a predetermined position consists of two steps: First, appropriate restriction fragments with the coding sequences are ligated . DNA is prepared in E.coli and subcloned in M13 to obtain single-stranded templates. In the next step, nucleotide sequences that are in excess are removed by in vitro mutagenesis. The precise in-frame boundary between the t-PA A chain and the u-PA B chain is at the activation site. Mutant DNA is subcloned into an appropriate expression vector for yeast and mammalian cell lines. a) Isolation of the DNA fragment encoding t-PA A-k. ieden: 10 pg of plasmid pJDB207/ PH05-I-TPA (see example 6) is cut with BamHI and PvuII. On 1.7. The kb BamHI-PvuII fragment is separated on a 0.8% agarose gel in Tris-borate-EDTA buffer pH 8.3. The DNA fragment contains the PH05 promoter, the invertase signal sequence and the coding sequence for finished t-PA up to the PvuII restriction site [cf. fig. 1; nucleotide positions 1305-1310]. DNA is electroeluted, precipitated with ethanol and resuspended in HgO at a concentration of 0.1 pmol/pl. b) Isolation of a DNA fragment encoding the u-PA B chain: Plasmid pCS16/UPA (see example 7B) is cut with Ball (cf.

figures 3 and 4, nucleotide positions 573-578) and Xhol. The 868 bp Ball-Xhol fragment is isolated as above and resuspended in HgO at a concentration of 0.1 pmol/µl.

c) Ligation of fragments to the vector fragment:

Plasmid pJDB207/PH05-TPA18 (European Patent Application No. 143,081)

is cut with BamHI and XhoI. The 6.7 kb vector fragment is isolated on a 0.85µ agarose gel in Tris-acetate buffer pH 8.2. DNA was electroeluted, ethanol precipitated and resuspended in HgO at a concentration of 0.1 pmol/pl.

0.2 pmol 1.7 kb BamHI-PvuII fragment, 0.2 pmol 868 bp Ball-XhoI fragment and 0.1 pmol 6.7 kb BamHI-XhoI vector fragment are ligated in 10 µl 60 mM Tris-HCl pH 7.5, 10 mM MgCl2, 5 mM DTT, 3.5 mM ATP and 400 units of T4 DNA ligase (Biolabs) at 15°C for 16 hours. One and 3 µl samples of the ligation mixture are added to 100 µl Ca<++> treated E.coli HB101 cells. The transformation is performed as usual.

Six transformed, ampicillin-resistant colonies are grown in LB medium containing 100 mg/l ampicillin. Plasmid DNA is prepared according to the method of Holmes et al.

[Analyst. Biochem. 114, 193 (1981)] and analyzed by restriction digestion with BamHI and Pstl. A clone with the expected restriction fragments is referred to as pJDB207/PH05-I-TPA<Å>UPA<B>.

Example 10: A u-PA/t-PA hybrid plasminogen activator with the u-PA A-k.ide domains and the t-PA B-k.ide (primary DNA construct p. i on)

The primary hybrid DNA construct consists of the u-PA nucleotide sequences from the SmaI site to the EcoRI site (see Fig. 4) attached to the t-PA nucleotide sequences from the Scal site (positions 940-945) to the XhoI site introduced by position 1800 via an Xhol linker. The resulting hybrid DNA sequence contains excess nucleotides that are removed by in vitro mutagenesis. The precise, in-frame boundary between the u-PA A chain and the t-PA B chain is at the activation site.

a) Isolation of a DNA fragment encoding u-PA A-k. 1eden: 7 pg of plasmid pCS16/UPA was cut with EcoRI. The sticky ends of the resulting 3 fragments are converted to blunt ends by a fill-in reaction with 7.5 units of Klenow DNA polymerase (BRL) in the presence of 60 mM Tris-HCl pH 7.5, 10 mM MgCl2, 0.1 mM dATP and 0.1 mM dTTP for 30 min. at 25°C. The reaction is stopped by adding EDTA to a final concentration of 12.5 mM. The DNA is further cut with Kpnl. A 619 bp KpnI butt [EcoRI] end fragment is isolated on a 1,556 agarose gel in Tris-borate-EDTA buffer pH 8.3, electroeluted and ethanol precipitated. b) Isolation of a DNA fragment encoding the t-PA B chain: 6 pg of plasmid pJDB207/PH05-TPA18 is cut with Seal and

Xhol. An 860 bp fragment is isolated on a 1,256 agarose gel in Tris-borate-EDTA buffer pH 8.3, electroeluted and ethanol precipitated.

c) Ligation of the DNA fragments into a pUC18 derived vector:

5 pg of plasmid pCS16/UPA (see example 7) is cut with

Kpnl and Xhol. The resulting 2.7 kb fragment is isolated on a 0.856 agarose gel in Tris-borate-EDTA buffer pH 8.3. DNA was electroeluted and ethanol precipitated. All the DNA fragments were resuspended in HgO at a concentration of 0.1 pmol/pl.

0.2 pmol of 619 bp Kpn-butt end u-PA fragment, 0.2 pmol 860 bp Scal-Xhol vector fragment was ligated as described above (Example 9). Ca<++> treated E.coli HB101 cells were transformed. 12 transformed, ampicillin-resistant colonies are grown in LB medium supplemented with ampicillin (100 mg/l). DNA is prepared according to Holmes et al.

(supra) and analyzed by restriction digestion with EcoRI and

Pst. A single clone with the expected restriction fragments is referred to as pCS16/UPA<A>TPA<B>.

Example 11: A u-PA/t-PA hybrid plasminogen activator with the second ring and the catalytic B chain of t-PA

(primary construction)

A hybrid plasminogen activator gene consisting of the DNA sequences of the urokinase "growth factor-like" (U) domain, the second kringle domain (K2) of t-PA, and the catalytic B chain of t-PA is constructed as follows: Two DNA restriction fragments encoding the u-PA growth factor domain and the t-PA K2 kringle and B chain, respectively, are ligated and inserted into plasmid pCS16. The resulting clone is named pCS16/UK2TPA<B>. A fragment containing the u-PA and t-PA coding sequences is subcloned into M13. In vitro mutagenesis is performed on single-stranded DNA to remove excess DNA sequences at the boundary between the u-PA and t-PA sequences.

5 pg of plasmid pCS16/UPA is cut with NcoI (nucleotide positions 326-331, see Fig. 4). The sticky ends of the restriction fragments are filled in a reaction with 5 units of Klenow DNA polymerase I (BRL) in the presence of 0.1 mM each of dATP, dTTP, dCTP, dGTP, 60 mM Tris-HCl pH 7.5, 10 mM MgCl2 in 50 pl for 30 min. at room temperature. The reaction is stopped by adding EDTA to a final concentration of 12.5 mM. DNA is ethanol precipitated and further cut with Xhol. The 3 kb Xhol blunt end [Ncol] fragment is isolated on a 0.8% agarose gel in Tris-borate-EDTA pH 8.3, electroeluted and ethanol precipitated. This fragment contains the pCS16 vector and the sequence encoding the u-PA growth factor domain. 10 pg of plasmid pJDB207/PH05-TPA18 (European Patent Application No. 143,081) is cut with BstXI [nucleotide positions 577-588]. The linear DNA fragment with 3' overhangs is incubated with 10 units of T4 DNA polymerase (BRL) in 100 µl of 33 mM Tris-acetate pH 7.9, 66 mM potassium acetate, 10 mM magnesium acetate, 0.5 mM DTT and 0.1 mg/ml bovine serum albumin for 2.5 min. at 37°C. The incubation is then continued for 35 min. at 37°C in the presence of 0.1 mM each of dATP, dCTP, dTTP, DGTP in a total volume of 200 µl. DNA is ethanol precipitated and cut further with Xhol. The 1.2 kb blunt end [BstXI]-Xhol fragment is separated on a 0.8% agarose gel, electroeluted and ethanol precipitated. This fragment contains the sequence coding for K2 and the B chain of t-PA.

0.2 pmol of the 1.2 kb t-PA fragment and 0.1 pmol of the 3 kb u-PA/vector fragment (see above) are ligated as described. Aliquots of the ligation mixture are used to transform competent E.coli HB101 cells. Ampicillin-resistant colonies are selected on LB agar plates containing 100 mg/l ampicillin. DNA is prepared from individual transformants and analyzed by Seal and Smal restriction cuts. A clone containing the 0.5 kb Seal and 1.55 kb Narrow border fragments is selected and referred to as pCS16/UK2TPA<B.>

Example 12: A t-PA/u-PA hybrid plasminogen activator with the second ring of t-PA and the catalytic B chain of u-PA (primary construction).

A hybrid plasminogen activator gene consisting of the DNA sequence of the urokinase "growth factor-like" (U) domain, the second ring (K2) of t-PA, and the catalytic B chain of u-PA is constructed in a manner analogous to that was described in example 11.

Construction of plasmid PCS16/UKQ UPAB-:

5 pg of plasmid pCS16/UPA is cut with BglII and NcoI (nucleotide positions 391-396 and 326-331, respectively, see Figure 4). The sticky ends of the restriction fragments are filled in a reaction with Klenow DNA polymerase I (BRL) as described above. The blunt-ended 4.2 kb DNA fragment is isolated on a 0.8% agarose gel in Tris-acetate buffer pH 8.2. DNA is electroeluted and ethanol precipitated. This fragment contains the u-PA G domain and the u-PA B chain attached to the vector molecule. 10 pg of plasmid <p>31/PH05-TPA18 (European Patent Application No. 143,081) is cut with Alul. A 447 bp Alul fragment containing the entire K2 domain of t-PA is isolated on a 1.5% agarose gel in Tris-borate-EDTA pH 8.3. The DNA fragment is electroeluted and ethanol precipitated.

0.2 pmol of the 477 bp fragment and 0.1 pmol of the 4.2 kb fragment are ligated. Aliquots of the incubation mixture are used to transform competent E.coli HB101 cells. Transformed cells are selected on LB agar plates with 100 mg/l ampicillin. DNA is prepared from ampicillin-resistant cells and analyzed by EcoRI and Seal cuts. A single clone showing the 551 bp EcoRI fragment and a 403 bp Seal fragment contains the Alul fragment in the correct orientation. This clone is referred to as pCS1b/UKgOT-k13.

Example 13: Cloning of primary hybrid DNA constructs in M13mpl8.

A) Cloning of pJDB207/ PH05- I- TPA^ UPA£ BamHI from<g>mented in M13mpl8

1.5 pg pJDB207/ PH05-I-TPAAUPAB (cf. Example 9) is obtained from a rapid DNA preparation is cut with 9 U BamHI (Boehringer) in 20 µl 10 mM Tris-HCl pH 7.5, 6 mM MgClg, 100 mM NaCl, 6 mM mercaptoethanol at 37 °C for one hour. After addition of 1 µl RNase (Serva, 1 mg/ml), incubation for 15 min. at 37°C and phenolization, the 2.5 kb insert is isolated on a 0.856 preparative agarose gel. DNA is extracted by electroelution and precipitated.

1 pg of M13mpl8 (RF) is cut with BamHI, treated with calf intestinal alkaline phosphatase and the 7.3 kb vector fragment is isolated on a 0.7 56 preparative agarose gel. DNA is electroeluted and precipitated. 100 pmol of M13mpl8 BamHI cut vector and 200 pmol of BamHI TPA<A>UPA<B> insert are ligated in 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400 U T4 DNA ligase for 7 hours at 15°C. After incubation at 65°C for 10 min., 5 µl of this ligation mixture is used for transformation of E.coli JM101 competent cells according to manual "M13 cloning and sequencing manual " published by Amersham. 36 colorless plaques were picked, and single-stranded and replicative form (RF) DNA is prepared. Analysis of the RF DNA shows that all clones contain inserts of the correct size after cutting with BamHI. Fragments of the correct size after cutting with EcoRI and Pstl indicates that the DNA insert in all clones is in the wrong orientation (single-stranded template DNA is the noncoding strand). One of these clones is referred to as mpl8/BamHI/TPA<A>UPA<B> and is used in deletion mutagenesis.

B. Cloning of a pCS16/ UPA^ TPÅB- Kpnl- Hindi 11 fragment in M13mpl8

1.5 pg pCSl6/UPA<A>TPA<B> (cf. Example 10) from a rapid DNA preparation is cut with 12 U KpnI in 20 µl 10 mM Tris-HCl pH 7.5, 6 mM MgCl2, 6 mM mercaptoethanol at 37 " C for one hour. After addition of 1 µl of 1 M NaCl, the DNA is cut with 12 U HindIII at 37°C for one hour. A 1.5 kb fragment is isolated on a 0.856 preparative agarose gel. DNA is extracted by electroelution and precipitated.

0.5 pg M13mpl8 (RF) is cut with KpnI and HindIII. The 7.3 kb vector fragment is isolated on a 0.756 preparative agarose gel. DNA is electroeluted and precipitated.

100 pmol of M13mpl8 Kpnl-HindiII cut vector and 200 pmol of Kpnl-Hindlll insert are ligated in 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400 TJ T4 DNA -ligase for 7 hours at 15°C. The reaction is stopped by incubation at 65°C for 10 min. 5 µl of this ligation mixture is used for transformation of E.coli JM101 competent cells. Ten colorless plaques are picked, and single-stranded and replicative form (RF) DNA is prepared. Analysis of the RF DNA shows that all clones contain inserts of the correct size and fragments of the correct size. One of these clones is referred to as mpl8/KpnI-HindIII/- UPA<A> TPA<B> and is used in deletion mutagenesis.

C. Cloning of a PCSI6/UK0 TPAS Kpn1-HindIII fragment in Ml3mpl8

1.5 pg pCS16/UK2TPA<B> (cf. Example 11) from a rapid DNA preparation is cut with 12 U Kpnl (Boehringer) in 20 µl 10 mM Tris-HCl pH 7.5, 6 mM MgCl2, 6 mM mercaptoethanol at 37° C for an hour. After addition of 1 µl of 1 M NaCl, the DNA is cut with 12 U of HindIII at 37°C for one hour. A 1.5 kb fragment is isolated on a 0.856 preparative agarose gel. DNA is extracted by electroelution and precipitated.

0.5 pg M13mpl8 (RF) is cut with KpnI and HindIII. The 7.3 kb vector fragment is isolated on a 0.756 preparative agarose gel. DNA is electroeluted and precipitated.

100 fmol of M13mpl8 Kpnl-HindiII cut vector and 200 fmol of Kpnl-Hindlll insert are ligated in 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400 U T4 DNA -ligase for 7 hours at 15°C. The reaction was stopped by incubation at 65°C for 10 min. 5 µl of the ligation mixture is used for transformation of E.coli JM101 competent cells. Seven colorless plaques are picked, and single-stranded and replicative form (RF) DNA is prepared. Analysis of RF DNA shows that all clones contain inserts of the correct size

and fragments of the correct size. One of these clones is referred to as mpl8/KpnI-HindIII/UK2TPA<B> and is used in deletion mutagenesis.

D. Cloning of a pCSlé/ TJKg TJPA^ Kpnl- Hindllll fragment in M13mpl8

1.5 pg pCS16/UK2UPA<B> (cf. Example 12) from a rapid DNA preparation is cut with 12 U KpnI in 20 µl 10 mM Tris-HCl pH 7.5, 6 mM MgCl2, 6 mM mercaptoethanol at 37°C in a hour. After addition of 1 µl of 1 M NaCl, the DNA is cut with 12 TJ of HindIII at 37°C for one hour. A 1.7 kb fragment is isolated on a 0.856 preparative agarose gel. DNA is extracted by electroelution and precipitated.

0.5 pg M13mpl8 (RF) is cut with KpnI and HindIII. The 7.3 kb vector fragment is isolated on a 0.756 preparative agarose gel. DNA is electroeluted and precipitated.

100 fmol of M13mpl8 Kpnl-Hindlll cut vector and 200 fmol of Kpnl-Hindlll insert are ligated in 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400 TJ T4 DNA -ligase for 7 hours at 15°C. The reaction is stopped by incubation at 65°C for 10 min. 5 µl of the ligation mixture is used for transformation of E.coli JM101 competent cells. Ten colorless plaques are picked, and single-stranded and replicative form (RF) DNA is prepared. Analysis of RF DNA shows that all clones contain inserts of the correct size and fragments of the correct size. One of these clones is referred to as mpl8/KpnI-HindIII/UK2U-PA<B> and is used in deletion mutagenesis.

Example 14: Deletion mutagenesis of primary hybrid DNA constructs. in oners

A) General protocol for sharing. ion mutagenesis.

a) Phosphorylation of mutagenic primer:

For mutagenesis, 200 pmol of the mutagenic primer is used

phosphorylated in 20 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 5 mM DTT, 0.1 mM spermidine, 0.1 mM EDTA containing 1 µl 10 mM ATP using 8 TJ T4 polynucleotide kinase (Boehringer, 8 TJ/pl ). After incubation at 37°C for one hour, the reaction is stopped by heating at 65°C for 10 min.

For hybridization screening, 20 pmol of the mutagenic primer is phosphorylated as above with 30 pCi of γ<32>P-ATP (3000 Ci/mmol; Amersham International) as the sole ATP source. The primer is diluted with 3.5 ml 6 x SSC and used immediately as a probe. b) Base pairing of mutagenic primer and universal sequencing primer to the single-chain template.

0.2 pmol of single-stranded template is incubated with 20 pmol phosphorylated mutagenic oligodeoxyribonucleotide primer (10 pmol/µl) and 10 pmol universal M13 sequencing primer in 10 µl 20 mM Tris-HCl pH 7.5, 10 mM MgCl2, 50 mM NaCl, 1 mM DTT at 95°C for 5 min. The solution is allowed to cool slowly to room temperature over a period of 30 min.

c) Extension 1igation reaction

To the above base-paired mixture is added 10 pl

enzyme-dNTP (dATP, dGTP, dTTP, DCTP) solution containing 1 µl buffer [0.2 M Tris-HCl pH 7.5, 0.1 MgCl2, 0.1 M DTT], 4 µl 2.5 mM dNTP mix, 1 µl 10 mM ATP, 0.5 µl T4 DNA ligase (Biolabs, 400 TJ/pl), 0.67 pl Klenow DNA polymerase (BRL, 2.99

U/jj1 ). The mixture is incubated at 15°C for one hour and then incubated at 8°C for 16 hours. The reaction is stopped by incubation at 65°C for 10 min.

d) Transformation of ligation mixture

The ligation mixture is diluted 1:20 and 1:200 with TE, 1 pl

and 5 µl of each of these diluted solutions is used to transform 0.2 ml of a non-repairing strain of E. coli BMH 71-18mutS [BMH71-18(A(lac-proAB), thi, supE. FllaciQ, ZAM15, <p>roA<+>B<+>l competent cells The construction of E.coli BMH71-18mutS (BMH71-18, mut S215::Tnio) is described by Kramer et al [Cell 38, 879-887 (1984 )].After the transfection, the basal cells are taken from the repair + strain of E.coli JM101 to minimize exposure of the phage to the mutant phenotype of the repair-minus strain [P. Carter, H. Bedouelle and G. Winter, Nucl. Acids Res. 13, 4431-4443 (1985 )].

e) Screening of the subjects

100 plaques that resulted from the transfected DNA become

picked and applied to YT plates and grown as colonies of infected bacteria for 15-18 hours. Colony blotting was adapted from Grunstein and Hogness [Proe. Nati. Acad. Pollock. USA 72, 3961-3965 (1985)]. A nitrocellulose filter (Millipore S.A., Cat. No. HAWP 090, pore size 0.45 pm) is placed on the colony plate for 10 min. at room temperature. The filters are denatured with 0.5 N NaOE, neutralized with 1 M Tris-HCl pH 7.5, and then treated with a high-salt solution (0.5 M Tris-HCl pH 7.5 + 1.5 M NaCl). The filters were baked in vacuum for 30 minutes at 80°C, prehybridized in 100 ml of 10x Denhardt's solution (D.T. Denhardt, Biochem. Biophys. Res. Commun. 23, 641-646), 6x SSC and 0.2% SDS in a sealable bar plastic bag for 15 minutes.

For the hybridization screen, the prehybridized filters were washed in 50 ml 6 x SSC for 1 minute and then hybridized in 3.5 ml probe containing <32>P-labeled mutagenic primer for 30 minutes. Eybridized filters are washed three times in 100 ml 6 x SSC at room temperature for a total of 2 minutes and then subjected to autoradiography. Good discrimination between wild-type and mutant phages is achieved by a quick wash (5 min) at 60°C in 100 ml 0.1 x SSC + 0.1% SDS.

f) Confirmation of shares. ion mutase. ion in positive clones after h<y>bridiserin<g>en

The phages from the positive clones are picked with toothpicks into 1 ml of 2 x YT, heated at 70°C for 20 minutes to kill the bacteria, and then 100 µl of this phage suspension was inoculated into 1.5 ml of a freshly grown E.coli JM101 culture (OD600 ~ 0.45). The cultures are shaken vigorously (300 rpm) at 37°C for 4 hours. Phage stock and DNA in replicative form from the positive clones are prepared [J. Messing, Methods in Enzymology, 101, 21-78 (1983)]. DNA from the mutants (after deletion mutagenesis) is analyzed with appropriate restriction enzymes and compared to the restriction fragments of wild-type (before deletion mutagenesis) DNA. After confirmation by restriction analysis, DNA from a correct mutant plaque is purified. Mutations are further confirmed by restriction analyzes and sequencing using the chain terminator method [F. Sanger, S. Niclen and A.R. Coulson, Proe. Nati. Acad. Pollock. USA 74, 5463-5467 (1977)]. B) Shared. ion mutagenesis on mp! 8/ BamHI/ TPA^ UPA5- (see Figure 23).

Deletion mutagenesis is performed as described in the general protocol. Positive clones from the hybridization are confirmed by restriction analysis. 333 bp are removed by deletion mutagenesis from the BamHI fragment. Restriction analysis with BamHI confirms the 2150 bp fragment. Further restriction analysis with EcoRI gives 660, 416, 287, 230 bp fragments on the mutants instead of the 660, 472, 416 and 287 fragments found in the wild type. Analysis with Pstl shows two fragments, 611 and 414 bp in size in the case of the mutants. Wild-type DNA shows three fragments of 622, 611 and 414 bp. A mutant clone having the correct structure is referred to as mpl8/BamHI/M0TPA<A>UPA<B>.

The DNA sequence at the boundary between the t-PA A chain and the u-PA B chain has been confirmed by the chain terminator sequencing method to have a sequencing primer of sequence

5' CAGAGCCCCCCGGTGC 3'.

This primer is complementary to the coding strand of u-PA (682-666).

C) Shared. ion mutagenesis on mpl8/ KpnI- HindIII/ UPAATPA^ ( see Figure 24 )

Deletion mutagenesis is performed as described in the general protocol. Positive clones from the hybridization are confirmed by restriction analysis with Pstl. In the mutants a 467 bp band is observed compared to the wild type which has a 544 bp fragment. A mutant clone having the correct structure is referred to as mpl8/KpnI-HindIII/M0UPA<A->TPA<B>. The deletion is confirmed by the chain terminator sequencing method using a sequencing primer with the sequence

5' CAAAGATGGCAGCCTGC 3'

This primer is complementary to the coding strand of t-PA (1062-1046).

D. Split ion mutagenesis on mpl8/ KpnI- HindIII/ IH^ TPA5- ( see Figure 25 )

Deletion mutagenesis is performed as described in the general protocol. Positive clones from the hybridization are confirmed by restriction analysis with KpnI-HindIII, Eco RI and Pstl. The fragments that emerge are

The correct size of the insert and the correct size of the fragments are observed in the mutants. A mutant clone having the correct structure is referred to as mpl8/KpnI-HindIII/- MOUKgTPA15. The deletion is confirmed by the chain-terminator sequencing method using a sequencing primer with the following sequence

5' CCCAGTGCCTGGGCACTGGGGTTCTGTGCTGTG 3'.

This primer is complementary to the encoded chain of t-PA (853-821).

E. Shared. ion mutagenesis on mpl8/ KpnI- HindIII/ UKg UPA5- ( see figure 261

Two separate deletion mutations are involved in the construction of UK2UPA<B>: First deletion mutagenesis is performed as described in the general protocol. Positive clones from the hybridization are confirmed by restriction analysis with EcoRI. In the mutants, 549, 416, 351 bp bands are observed, in contrast to the wild type, which has the following fragments: 549, 452 and 416 bp. A mutant clone having the correct structure is referred to as mpl8/KpnI-EindIII/M0UK2UPA<B->1. The deletion is confirmed by the chain-terminator sequencing method using a sequencing primer with the following sequence:

5' CCCAGTGCCTGGGCACTGGGGTTCTGTGCTGTG 3'.

The primer is complementary to the encoded chain of t-PA (853-821).

In the second step of deletion mutagenesis, a deletion is made simultaneously with the introduction of a point mutation. The deletion mutagenesis is performed as described in the general protocol. Positive clones from the hybridization are confirmed by restriction analysis with EcoRI. In the mutants, 416, 351, 259 bp bands are observed in contrast to the wild type, which has 549, 416 and 351 bp fragments. A mutant clone having the correct structure is referred to as mpl8/KpnI-HindIII/M0UK2UPA<B.> The deletions are confirmed by the chain terminator sequencing method using a sequencing primer with the following sequence

5' CAGAGCCCCCCGGTGC 3'.

The primer is complementary to the encoded chain of u-PA (682-666).

Example 15: Cloning of the hybrid t-PA/u-PA cDNA constructs into yeast expression vector pJDB207

A) Cloning of the TPaAuPA^ hybrid gene into pJDB207

RF DNA was prepared for mpl8/BamHI/M0TPA<A>UPA<B> using the rapid DNA isolation procedure [D.S. Holmes and M. Kingley, Anal. Biochem. 114, 192-197 (1981)].

RF DNA (~1.5 pg) is cut with 9 U BamHI in 20 µl 10 mM Tris-HCl pH 7.5, 6 mM MgCl 2 , 100 mM NaCl, 6 mM mercaptoethanol for one hour at 37°C. After addition of 1 µl of RNase (1 mg/ml) and incubation for 10 minutes at 37°C, the 2.1 kb insert is isolated on a 0.7% preparative agarose gel. The DNA deposit is extracted by electroelution and precipitation in ethanol.

1.5 pg of pJDB207/ PH05-I-TPAAUPAB is cut with BamHI, treated with calf intestinal alkaline phosphatase and the 6.7 kb vector is isolated. After electroelution, the vector DNA is precipitated.

100 fmol of pJDB207/ PHO5-1-TPAAUPAB BamHI cut vector, 200 fmol TPA<B>UPA<B> insert is ligated in 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400

TJ T4 DNA ligase for 8 hours at 15°C. The reaction is stopped by incubation at 65°C for 10 minutes. 5 µl of this ligation mixture is used for transformation of E.coli HB101 Ca<2+ >cells [M. Dagert and S.D. Ehrlich, Gene 6, 23-28 (1979)]. 12 amp<E> colonies are picked and DNA is prepared by the rapid isolation procedure. After analysis of the DNA, 5 clones show both the correct size of the insert and the correct orientation. A clone is grown in 100 ml LB medium containing 100 mg/ml ampicillin. Plasmid DNA is isolated and is referred to as pJDB207/PH05-I-MOTPA<A>UPA<B>.

B) Cloning of MOUPaAt<p>aI. the aiOTJK2 TPAB and MOUKo UPAB gene inserts into plasmid pCS16

RF DNA is prepared for mpl8/KpnI-HindIII/M0UPA<A>TPA<B>» mpl8/KpnI -Hindi 11 /M0UK2TPAB , mpl8/KpnI -Hind 11 /M0TJK2UPAB using the rapid DNA isolation procedure.

The three RF-DNA (~1.5 pg) each cut with 12 U Kpn1 and 12 U HindIII in 20 µl 10 mM Tris-HCl pH 7.5, 6 mM MgCl2, 6 mM mercaptoethanol for one hour at 37°C. 1 µl 1 M NaCl is added and the DNA is further cut with 12 TJ HindIII. After addition of 1 µl RNase (1 mg/ml) and incubation for 10 min. at 37°C, the 1.4 kb insert is isolated on a 0.8% preparative agarose gel. The DNA deposit is extracted by electroelution and precipitated in ethanol.

Three pg of pCS16/UPA are cut with KpnI and HindIII and the 2.7 vector fragment is isolated. After electroelution, vector DNA is precipitated in ethanol.

100 fmol of pCS16 KpnI-HindIII cut vector, 200 fmol of KpnI-HindIII cut insert fragments are ligated in 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400 TJ T4 DNA ligase for 8 hours at 15°C. The reaction is stopped by incubation at 65°C for 10 minutes, 5

µl of this ligation mixture is used for transformation of E.coli HB101 Ca<2+> cells.

Six amp<R> colonies are picked for each of the three ligations. DNA is prepared by the rapid isolation procedure. After analysis of the DNA with KpnI-HindIII, insert bands of the correct size were observed. One clone of each of the three ligations was grown in 100 ml LB medium containing 100 µg/ml ampicillin. Plasmid DNA derived from mpl8/KpnI-HindIII/M0UPA<A>TPA<B>, mpl8/KpnI-HindIII/M0UK2TPA<B> and mpl8/KpnI-HindIII/M0UK2UPA<B> are isolated and are referred to as pCS16/M0TJPAATPAB , pCS16/M0UK2TPA<B> and pCS16/M0UK2UPA<B>, respectively. C) Cloning of the M0UPAATPA<B>, M0TJK2TPAB and M0TJK2UPAB gene inserts into pJDB207.

Five pg of PJDB207/ PH05-I-UPA are cut with 15 U Seal and 15 TJ Xhol (Boehringer) in 50 µl 10 mM Tris-HCl pH 7.5, 6 mM MgCl2, 150 mM NaCl, 6 mM mercaptoethanol for one hour at 37° C. After addition of 1 µl RNase (1 mg/ml), the 6.7 kb vector fragment is isolated. After electroelution, vector DNA is precipitated.

Fifteen pg each of pCS16/M0TJPAATPAB, pCS16/M0UK2TPA<B>, pCS16/M0TJK2UPAB are incubated at 37°C for one hour with 30 TJ Xhol in 200 10 mM Tris-HCl pH 8, 6 mM MgCl2, 150 mM NaCl, 6 mM mercaptoethanol, extracted with an equal volume of phenol-chloroform and precipitated in ethanol. The precipitated Xhol cut pCS16/M0TJPA<A>TPA<B>, pCS16/M0TJK2TPAB and pCS16/M0UK2UPA<B >DNA are each resuspended in 150 µl of 10 mM Tris-HCl pH 7.5, 6 mM MgCl2, 150 mM NaCl. 6 mM mercaptoethanol, 1.5 pg ethidium bromide, incubated at 37 °C for 40 min with 12 TJ Seal (partial cut), and extracted with an equal volume of phenol, followed by an equal volume of chloroform-isoamyl alcohol (50:1 ). The 1.2 kb fragments are each isolated on a 156 preparative agarose gel. DNA is extracted by electroelution and precipitated.

100 fmol of pJDB207/ PHO5-1-UPA Scal-Xhol cut vector and 200 fmol of Xho-Scal cut pCS16/M0UPA<A>TPA<B>, pCS16/M0UK2TPA<B >or pCS16/M0UK2UPA<B> 1.2 kb inserts , respectively, are ligated in 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400 TJ T4 DNA ligase for 16 hours at 15° C. The reaction is stopped by incubation at 65°C for 10 minutes. 5 µl of this ligation mixture is used for transformation of E.coli HB101 Ca<2+> cells.

Six amp<R> colonies are picked from each of the three ligations. DNA is prepared using the rapid isolation procedure. Restriction analysis of the DNA shows bands with the correct size of the insert. A clone from each of the three ligations is grown in 100 ml LB medium containing 100 µg/ml ampicillin. Plasmid DNA derived from pCS16/M0UPA<A>TPA<B>, pCS16/M0TJK2TPAB, pCS16/M0TJK2UPA<B> is referred to as pJDB207/PH05-I-M0UPA<A>TPA<B>, pJDB207/PH05-I- M0UK2TPA<B> and pJDB207/PH05-I-M0TJK2UPA<B>, respectively.

Example 16: Transformation of Saccharomyces cerevisiae GRF18 and production of yeast cell extracts

The plasmids pJDB207/PH05-I-MOTPAAUPAB , pJDB207/ PHO5-1 -M0TJPAA-TPA<B>, pJDB207/PH05-I-MOTJK2TPA<B> and pJDB207/PH05-I-MOTJK2UPA<B> have each been introduced into Saccharomyces cerevisiae strain GRF18 using the transformation protocol as described by Hinnen et al. [Pro. Nati. Acad. Pollock. US 75, 1929 (1978)]. Five µg each of plasmid DNA is added to 100 µl of a spheroplast suspension and the mixture is treated with polyethylene glycol. The spheroplasts are mixed with 10 ml regeneration agar and plated out on yeast minimal medium plates without leucine. After incubation for 3 days at 30°C, approximately 200 transformed cells are obtained.

One colony from each of the yeast transformations is picked. The different colonies are referred to as

Saccharom<y>ces cerevisiae GRF18/pJDB207/PH05-I-MOTPA<Å>UPA<B >Saccharomyces cerevisiae GRF18/pJDB207/ PH05-I-MOUPAATPAB Saccharom<y>ces cerevisiae GRF18/pJDB207/m)5-I-MOUPAATPAB Saccharom<y>ces cerevisiae GRF18/pJDB207/PH05-I-M0UK2UPA<B>

Yeast cells are grown at 30" C in 20 ml HE-17 medium (8.4 g yeast nitrogen base (Difco), 10 g L-aspargine (Sigma), 1 g L-histidine (Sigma), 40 ml 5056 glucose per 1 liter solution) in a 50 ml Erlenmeyer flask with shaking for 24 hours until a density of 8-10x10<7> cells/ml is reached. The cells are centrifuged, resuspended in 10 ml of 0.956 NaCl. Two ml of the resuspended cells are used to inoculate 50 ml of low -P-^ minimal medium (as described in European patent application no. 143081) to which is added 10 g/l L-asparagine (Sigma), and 10 g/l L-histidine (Sigma), in 250 ml Erlenmeyer flasks. The incubation takes place at 30 °C at 250 rpm.

Cells from 10 ml of low V^ minimal medium are collected after 48 hours by centrifugation at 3000 rpm for 10 minutes in Falcon 2070 tubes. The cells are washed once with 10 ml of low P^ medium and centrifuged. The cell pellet is suspended in lysis buffer [66 mM potassium phosphate pH 7.4, 4 mM Zwittergent (Calbiochem.)]. 8 g of glass beads (0.5-0.75 mm in diameter) and a small glass rod are added to the cell suspension and the suspension is shaken on a Vortex Mixer (Scientific Instruments Inc., USA) at full speed for 4x2 min. at intervals of 2 min. on ice. More than 9056 of the cells are crushed by this procedure. Cell debris and the glass beads are sedimented by centrifugation for 5 min. at 3000 rpm at 4°C. The supernatant is used for the determination of PA activity and for the purification and isolation of PA. Example 17: Insertion of hybrid PA coding sequences into mammalian cell express. in ons vector A) Insertion of UPA<A>TPA<B> 'perfect' hybrid coding sequence.

RF DNA of mpl8/KpnI-HindIII/M0UPA<A>TPA<B> is cut at the Smal site located just upstream of the beginning of the coding sequence and is ligated to a SacI linker (CGAGCTCG). Next, the plasmid is cut with SacI, which cuts at the position of the ligated linkers and at the natural SacI site in the t-PA-derived part of the hybrid PA coding sequence. The smaller of the two resulting fragments is purified via an agarose gel and ligated to Sac1-cut pCGA44 (see Example 4), transformed into E.coli HB101 and DNA from candidate clones tested with EcoRI. A clone with the expected restriction pattern is referred to as pCGCl/-

UPA<A>TPA<B.>

B) Insertion of TJK2TPAB hybrid coding sequence.

RF DNA of mpl8/KpnI-HindIII/M0UK2TPA<B> is cut at the SmaI site located just upstream of the beginning of the coding sequence and ligated to SacI as above. After cutting with SacI, the resulting small fragment is cloned into SacI-cut pCGA44 as described above and a clone with the expected restriction pattern is referred to as pCGC2/UK2TPA<B>.

C) Insertion of UK2UPA<B> hybrid coding sequence.

RF DNA mpl8/KpnI-HindIII/M0UK2UPA<B> is cut at the SmaI site upstream of the u-PA coding sequence and at the XhoI site downstream of the coding sequence (in vector DNA). The sticky ends of the DNA fragment are filled using E.coli DNA polymerase I (cf. example 5D). SacI linkers are ligated to the blunt ends, DNA is cut with SacI, the smaller of the two resulting fragments is purified via an agarose gel and cloned into SacI-cut pCGA44. A clone with the expected EcoRI restriction pattern is referred to as pCGC3/-

UK2UPAB.

D) Insertion of a TPA<A>UPA<B> 'perfect' coding sequence.

Step 1: RF DNA of mpl8/BamHI/M0TPA<A>UPA<B> is cut with BamHI and the smallest (~2.1 kb) fragment is cloned into BamHI cut pJDB207/ PH05-I-TPAAUPAB (cf. example 9) vector. The correct orientation is selected by cutting with HindIII and a correct plasmid is called pJDB207/PH05-I-MOTPA<A>UPA<B.>

Step 2: A ~600 bp Sacl-Narl fragment from ptNC.UC (cf. example 3) and a -1350 bp Narl-Xhol fragment from pJDB207/-PH05-I-TPA<A>UPA<B> are isolated and cloned into the Sac1-Xhol cut pCS16 (cf. example 7) vector. The ~1.9 kb insert is confirmed by cutting with Sac1-XhoI and EcoRI. A correct plasmid is called pCS16/MOTPA<A>UPA<B>.

Step 3: Plasmid pCS16/M0TPA<A>UPA<B> is cut at the XhoI site located downstream of the u-PA coding sequence and the sticky ends are filled using E.coli DNA polymerase I. SacI linkers are ligated to the blunt ends and DNA are cut with SacI. The smaller of the two fragments is purified via an agarose gel and cloned into the Sac1-cut pBR4a (cf. example 5) vector fragment. Correct orientation and size of the insert are confirmed by cutting with BamHI and SacI, respectively. A proper plasmid is designated pCGC4a/-

TPA<A>UPA<B.>

Example 18: Construction of several hybrid PA coding sequences and insertion thereof into mammalian cell express. ion vector A) Cloning of a pCGC4a/TPA<A>UPA<B> fragment in M13mpl8. 3 pg of pCGC4a/TPA<A>UPA<B> (cf. Example 17) is cut with 12 U SacI (Boehringer) in 20 µl 10 mM Tris-HCl pH 7.5, 6 mM MgCl2, 6 mM mercaptoethanol at 37°C for one hour. A ~1.9 kb fragment is isolated on a 0.7% preparative agarose gel. DNA is extracted by electroelution and precipitated.

0.5 µg of M13mpl8 (RF) is cut with SacI. The 7.3 kb vector fragment is isolated on a 0.756 preparative agarose gel. DNA is electroeluted and precipitated.

100 fmol of M13mpl8 SacI cut vector and 200 fmol of SacI insert are ligated in 10 µl 50 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM DTT, 2 mM ATP, 0.5 pg gelatin with 400 U T4 DNA ligase for 7 hours at 15°C. The reaction is stopped by incubation at 65° C. for 10 min. 5 µl of this ligation mixture is used for transformation of E.coli JM101 competent cells. Six colorless plaques are picked, and single-stranded and replicative form (RF) DNA is prepared. After analysis of the RF DNA, four clones show the correct size of the insert and the correct orientation. One of these clones is referred to as mpl8/SacI/TPA<A>UPA<B> (BC).

B) Cloning of a pBR4a SacI fragment into M13mpl8.

A pBR4a (cf. example 5) SacI fragment is cloned into M13mpl8. One of the clones that has an insert of the correct size and orientation is referred to as mpl8/Sad/TPAAUPAB (BR).

C) Deletion mutagenesis on TPA-UPA hybrid constructs.

1) Construction of K2UPA<B> (BC) [i.e. tPA(1-3)-tPA(176-275)-uPA(159-411)]

Deletion mutagenesis is performed as described in the general protocol (cf. Example 14) on mpl8/SacI/TPA<A>UPA<B> (BC). Positive clones from the hybridization are confirmed by restriction analysis with SacI. In the mutants, a ~1380 bp band is observed compared to wild type which has a -1900 bp fragment. Mutants are further confirmed by cutting with EcoRI. A mutant clone having the correct structure is referred to as mpl8/SacI/K2UPA<B> (BC). The deletion is verified by the chain terminator sequencing method using a sequencing primer with the following sequence:

5' CCCAGTGCCTGGGCATTGGGGTTCTGTGCTGTG 3'

This primer is complementary to the coding strand of t-PA(853-821) with a mispaired base at position 838 (t-PA).

2) Construction of FUPA<B> (BC) [i.e. tPA(1-49)-tPA(262-275)-uPA(159-411)]

Deletion mutagenesis is performed as described in the general protocol (cf. Example 14) on mpl8/SacI/TPA<A>UPA<B> (BC). Positive clones from the hybridization are confirmed by restriction analysis with SacI. In the mutants, a ~1200 bp band is observed in contrast to wild-type which has a ~1900 bp fragment. The mutants are further confirmed with EcoRI cutting. A mutant clone having the correct structure is referred to as mpl8/SacI/FUPA<B> (BC). The deletion is verified by the chain terminator sequencing method using a sequencing primer with the following sequence

S' CAGAGCCCCCCGGTGC 3'.

This primer is complementary to the coding strand of u-PA (666-682).

3) Construction of FK2UPA<B> (BC) [i.e. tPA(1-49)-tPA(176-275)-uPA(159-411)]

Deletion mutagenesis is performed as described in the general protocol (cf. Example 14) on mpl8/SacI/TPA<A>UPA<B> (BC). Positive clones from the hybridization are confirmed by restriction analysis with SacI. In the mutants, a ~1470 bp band is observed compared to wild-type which gives a ~1900 bp fragment. Mutants are further confirmed by EcoRI cutting. A mutant clone having the correct structure is referred to as mpl8/SacI/KF2UPA<B> (BC). The deletion is verified by the chain-terminator sequencing method using a sequencing primer with the following sequence

5' CCCAGTGCCTGGGCATTGGGGTTCTGTGCTGTG 3'.

This primer is complementary to the coding strand of t-PA(853-821) with a base mismatch at position 838 (t-PA).

4) Construction of FGK2UPA<B> (BC) [i.e. tPA(1-86)-tPA(176-275)-uPA(159-411)]

Deletion mutagenesis is performed as described in the general protocol (cf. Example 14) on mpl8/Sad/TPA<A>UPA<B> (BC). Positive clones from the hybridization are confirmed by restriction analysis with SacI. In the mutants, a ~1580 bp band is observed compared to wild-type which gives a ~1900 bp fragment. The mutants are further confirmed by EcoRI cutting. A mutant clone having the correct structure is referred to as mpl8/SacI/FGK2UPA<B> (BC). The deletion is verified by the chain terminator sequencing method using a sequencing primer with the following sequence

5' CCCAGTGCCTGGGCATTGGGGTTCTGTGCTGTG 3'.

This primer is complementary to the coding strand of t-PA(853-821) with a mispaired base at position 838 (t-PA).

5) Similar deletion mutagenesis protocols are used to generate

K2UPA<B>(BR) [tPA(1-3)-tPA(176-262)-uPA(132-411)] FUPA<B>(BR) [tPA(1-49)-uPA(134-411) ]

FK2UPA<B>(BR) [tPA(l-49)-tPA(176-262)-uPA(132-411)] and FGK2UPA<B>(BR) [tPA(l-86)-tPA(176-262 )-uPA(132-411)]. D) Insertion of hybrid PA-encoded sequences into mammalian cell expression vector

1. Insertion of FUPA<B>(BC), K2UPA<B>(BC), FK2UPA<B>(BC) and

FGK2UPA<B>(BC).

RF DNA from mpl8/Sad/K2UPAB(BC ), mpl8/SacI/FUPA<B>(BC), mpl8/SacI/FK2UPA<B>(BC) and mpl8/SacI/FGK2UPA<B>(BC) are each cut with SacI. The smaller of the two resulting fragments is isolated and is ligated to the SacI cut pBR4a (cf. example 5) vector fragment. Transferred to E.coli HB101 and the correct orientation and size of the insert confirmed by cutting with BamHI and SacI, respectively. The resulting plasmids are designated pCGC5/K2UPA<B>, pCGC6/FUPA<B>, pCGC7/FK2UPA<B> and pCGC8/FGK2UPA<B>, respectively. 2. Likewise, K2UPA<B>(BR), FUPA<B>(BR), FK2UPA<B>(BR) and FGK2UPA<B>(BR) DNA (see above) are also inserted into pBR4a. The resulting plasmids are designated pBR5, pBR6, pBR7 and pBR8, respectively.

Example 19: Mammalian expression vectors containing the DHFR gene

Plasmid pSV2dhfr (ATCC 37145) is a plasmid that allows selection of transformants of DHFR-containing cells by selection using the antifolate drug methotrexate or selection of DHFR<+> transformants into DHFR" CHO cells [DUKXB] cells; G. Urlaub, Proe. Nat. Acad. Sei. U.S.A. 77, 4216-4220 (1980)]. Into the single BamHI site of this plasmid, the BamHI fragment of pCGA28 containing the modular t-PA gene can be cloned. Plasmids containing any preferably of the possible orientations are designated pCGA700a/tPA and pCGA700b/tPA Both can be used to express t-PA in tissue culture cells but pCGA700a/tPA is preferred, where the transcription of the t-PA gene is in the same direction as in the DHFR gene, as this orientation often leads to a somewhat higher degree of expression compared to plasmids that are convergently transcribed.

In an analogous manner, the modular genes encoding hybrid plasminogen activators (above) from the plasmids pBR1a/tPA, pBR2a/UPA<A>TPA<B>, pCGCl/UPA<A>TPA<B>, and pCGC2/UK2TPA<B > be combined as the BamHI fragments with the DHFR gene of pCGA700a/tPA to form the plasmids pCGA701a/tPA, pCGA702a/UPA<A>TPA<B>, pCGA705a/UPA<A>TPA<B>, and pCGA707a/UK2TPA<B > respectively, where the modular plasminogen activator gene is transcribed in the same direction as the DHFR gene, and pCGA701b/tPA, pCGA702b/UPA<A>TPA<B>, pCGA705b/UPA<A>TPA<B>, pCGA707b/UK2TPA< B> where both genes are transcribed in opposite directions. Due to the presence of a BamHI sequence in the part coding for the u-PA B chain, the modular plasminogen activator gene can only be isolated by a partial cut (2 out of 3 BamHI sites) of the neo^ plasmid followed by isolation of the the appropriate fragments (cf. figures) by agarose gel electrophoresis. Thus, from pBR3a/uPA, pBR4a/TPA<A>UPA<B>, pBr5/K2UPA<B>, pBR6/FUPA<B>, pBR7/FK2UPA<B>, pBR8/FGK2UPA<B>, pCGC3/UK2UPA< B>, pCGC4a/TPA<A>UPA<B>, pCGC5/K2UPA<B>, pCGC6/FUPA<B>, pCGC7/FK2<U>PAB, and pCGC8/FGK2UPA<B> can be constructed pCGA703a/uPA, pCGA704a/TPA<A>UPA<B>, pCGA705a/K2UPA<B>, pCGA708a/FUPA<B>, pCGA706a/FK2UPA<B>, pCGA707a/FGK2UPA<B>, pCGA709a/UK2UPA<B>, pCGA711a/TPA<B> A>UPA<B>, pCGA712a/K2UPA<B>, pCGA713a/FUPA<B>, pCGA714a/FK2UPA<B>, and pCGA715a/FGK2UPA<B>, respectively, where all the plasminogen activator genes are transcribed in the same direction as DHFR -gene, and further pCGA703b/uPA, pCGA704b/TPA<A>UPA<B>, pCGA708b/FUPA<B>, pCGA705b/K2UPA<B>, pCGA706b/FK2UPA<B>, pCGA707b/FGK2UPA<B>, pCGA709b/ UK2UPA<B>, pCGA711b/TPA<A>UPA<B>, pCGA712b/K2UPAB, pCGA713b/FUPA<B>, pCGA714b/FK2UPA<B> and pCGA715b/FGK2UPA<B>, where both genes are transcribed on a non- convergent way.

Example 20: Production of hybrid plasminogen activators by h. ielp of transformed mammalian cells A) Maintenance and DNA transfection of tissue culture cells; general procedure.

DNA constructs are expressed in DUKXB1, a mutant of Chinese hamster ovary (CHO) cells lacking the enzyme dihydrofolate reductase [G. Urlaub et al., Proe. Nat. Acad. Pollock. USA 77, 4216-4220 (1980)]. The DUKXB1 cells are grown in alpha-MEM medium containing nucleosides (GIBCO) supplemented with 5% fetal calf serum.

The cells are plated at a density of 10,000/cm in 6-well multiplates (3.4 cm diameter) and transformed with 4 pg DNA: DNA is dissolved at 50 pg/ml in 10 mM Tris-HCl pH 7.0 containing 0.1 mM EDTA, cooled on ice for .5 min., then 0.25 volume of 1 M CaCl2 is added and this is incubated on ice for 10 min. This mixture is then mixed with an equal volume of 2 x HBS (50 mM Hepes, 280 mM NaCl, 0.75 mM Na2HPO4, 0.75 mM NaH2PO4. pH 7.12) followed by 10 min. incubation on ice. Then this DNA-Ca-phosphate coprecipitate is added to the culture medium and the cells are incubated with DNA for 16-18 h, followed by a glycerol shock, i.e. the cells are rinsed with TBS (80 g/l NaCl, 3.8 g/l KC1, 1 g/l Na2HP04.2H20, 0.114 g/l CaCl2.2H20, 0.11 g/l MgCl2.6H20, 25 mM Tris/Hel pH 7.5) , incubated 1 min. with 20% (v/v) glycerol in TBS, rinsed again with TBS and cultured for 24 h in tissue culture medium. The cells are then trypsinized and transferred to 8 cm diameter Petri dishes. The next day, the original culture medium without selective agent is replaced with medium containing 1 mg/ml geneticin. The medium is replaced every three or four days. Colonies can be seen at approximately day 14. Cells from individual colonies are isolated by scraping them off with the tip of a pipette while simultaneously pulling them into the tip filled with a trypsin solution and then each is transferred to a well of a 24 well plate multi dishes that have been supplemented with medium containing geneticin. When these are full, the cultures are moved into the wells of a 6-well multi dish and then into 8 cm diameter Petri dishes.

B. Agarose plate assays for plasminogen activators.

These sensitive assays for plasminogen activators use agarose gels spiked with plasminogen (stock solutions made by dissolving plasminogen Sigma A-6877 at 1 mg/ml and then dialyzing twice against 100 volumes of 50 mM Tris/ECl pH 8.0) or either casein (spiked such as non-fat milk) or fibrin (added as fibrinogen plus thrombin). The sample containing plasminogen activator is activated into wells cast into a 4 mM thick agarose layer and the gel is then incubated at 37°C. The enzyme activity is then detected by the plasminogen activator diffusing radially away from the sample well, and transferring plasminogen in the gel to plasmin which in turn breaks down casein or fibrin and thus produces a clear ring in the opaque gel around the sample well. The radius of the ring (measured from the edge of the sample well) is a measure of the amount of activated plasminogen. The assay does not show a linear response to the amount of plasminogen activator added. For assays of low amounts of plasminogen activator, the incubation can be extended to several days. The procedure and calibration of the casein assay are described in Tang et al. [Ann. NEW. Acad. Pollock. 434, 536-540

(1984)] with the exception that instead of 2% (w/v) Carnation non-fat milk powder, 12.55é (v/v) sterilized (UHT) fat-free milk from Migros Corp. (Switzerland) used. When fibrin

[Granelli-Piperno and Reich, J. Exp. Med. 148, 223-234 (1978)]

is used as substrate, 0.2 g of agarose is dissolved in 15 ml of 0.9$ NaCl and cooled to 42°C. At this time, 5 ml of 0.9$ NaCl containing 80 mg of bovine fibrinogen (Sigma F-8630), 0.1 ml of plasminogen solution (above) and 0.1 ml of 100 mg/ml sodium azide are added at 42°C. Finally, 0.2 bovine thrombin (Sigma T-6634, dissolved at 16.6 NIH units/ml in 0.9% NaCl) is added and the solution is quickly poured into a Petri dish (8 cm

diameter), and this is cooled to room temperature for one hour. The resulting gel is approximately 4 mm thick and can be stored at 4°C for several days or used immediately in the same manner as the casein-containing gel described above.

C. Production of hybrid PA proteins in hamster cells.

CEO DUKXB1 cells are transformed with DNA from plasmids PBR1A, pBRlB, pBR2A, pBR2B, pBR3A, pBR3B, pBR4A, pBR5, pBR7, pBR8, pCGCl, pCGC2, pCGC3, pCGC4a, pCGC5, pCGC6, pCGC7 and pCGC8, respectively, as described above (Example 20A). Colonies are visible around day 10, and are picked around day 15 as described above and two weeks later the cell count is high enough for the PA values to be measured as described above. Non-transformed cells and cell lines transformed with pBR1B, pBR2B, pBR3B, which contain the inserted SacI fragment in the wrong (antisense) orientation do not produce detectable amounts of PA. D. Enzyme activity in media conditioned with transformed CEO cells.

Conditioned medium from plasmid transformed and control CEO cells is prepared by culturing 200,000-500,000 cells/ml for 24 hours in Alpha-MEM with nucleosides and 5% fetal calf serum and 0.03 ml is incubated on agarose plates containing casein or fibrin for the time period that are mentioned below. On the fibrin plate, a minimal background activity is detected, presumably due to endogenous hamster t-PA, in the DUKXB1-conditioned medium. No ring appears on the casein plates if the hybrid protein samples are mixed with 3 microliters of rabbit anti-tPA antibodies (made against purified Bowes melanoma (t-PA)) or anti-urokinase antibodies (made against Serono urokinase).

Anti-tPA antibody does not inhibit u-PA enzyme, nor does anti-urokinase antibody inhibit t-PA to an appreciable extent. The results are summarized in Table 1.

Example 21: Production of hvbridoma cells and isolation of monoclonal antibodies a) Immuno<g>en source: A sample of semi-purified natural human (melanoma t-PA) which has an assumed purity of >90#. b) Immunization protocol: Three groups of BALB/c mice (Tierfarm Sisseln, Switzerland) which are 10-14 weeks old are immunized by injection into the two hind footpads and subcutaneously 100 pg melanoma t-PA emulsified in complete Freund's adjuvant (Difco). Then, the first group (No. 405) receives 10 pg of t-PA in incomplete adjuvant, every week for six weeks, while the second group (406) receives the same amount every two weeks. The third group (407 ) receives twice 50 pg of t-PA at intervals of three weeks. All animals are euthanized at week 4 and week 8. At the last injection, 100 pg t-PA in PBS i.p. and four days later spleen cells are fused with SP2/o myeloma line according to standard procedure. Only those mice with a high anti-t-PA antibody titer are used in the fusion.

c) Cell fusion. ionization: All fusion experiments are performed according to the procedures of G. Kohler and C. Milstein

[ Nature 256, 495 (1975)] using the non-secreting Sp 2/0-Agl4 myeloma line [p. Shulman, CD. Wilde and G. Kohler, Nature 276, 269 (1978)]. 10<8> spleen cells are mixed with 10<7> myeloma cells in the presence of 1 ml 5056 polyethylene glycol (PEG 1500, Serva). After washing, the cells are resuspended in 48 ml of standard Dulbecco's minimum essential medium (Gibco No. 042501). 3x10^ normal mouse peritoneal aspirated cells per fusion are added as feeder cells. The cells are distributed into 48x1 ml costar wells and fed 3 times per week with standard BAT selection medium for 3 to 6 weeks. When the growth of hybridoma cells becomes visible, the supernatants are screened by both direct antigen-binding (ELISA) and neutralization (casein) assays (see below). The results of 4 fusion experiments are as follows: Out of 192 seeded wells, 192 hybridomas were obtained. Of these, 24 produced anti-t-PA antibody. Of 24 positive hybridomas, 14 were cloned and out of 574 clones, 31 produced stable anti-t-PA mAb. Three (clones 405B.33.3, 406A.23.7 and 407A.15.27) of these were injected into mice and ascites fluid has been produced for further studies.

d) Isolation and purification of monoclonal antibodies:

BALB/c mice 8-10 weeks old (Tierfarm Sisseln, Switzerland) become

pretreated intraperitoneally with 0.3 ml pristane (Aldrich). 2-3 weeks later, 2-5xl0<6> hybridoma cells were cloned

405B.33.3, 406A.23.7 and 407A.15.27 and 0.2 ml pristane inoculated intraperally in tonealt. After 8-10 days ascites fluid collected, and centrifuged at 800 x g and stored at -20°C.

Thawed ascites fluid is centrifuged at 50,000 x g for 60 min. A fat layer floating on the surface is carefully removed, and the protein concentration is adjusted to a concentration of 10-12 mg/ml. Crude immunoglobulin is precipitated by the dropwise addition of 0.9 volume equivalents of saturated ammonium sulfate at 0°C, then dissolved in 20 mM Tris-HCl/50 mM NaCl (pH 7.9) and dialyzed against the same buffer. An immunoglobulin fraction is obtained by DEAE-D52 cellulose (Whatman) chromatography where a buffer gradient system with 20 mM Tris-HCl/25-400 mM NaCl, pH 7.9 is used. The immunoglobulin is again precipitated with ammonium sulphate and dissolved in PBS at a concentration of 10 mg/ml.

Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) of monoclonal antibodies demonstrates a degree of purity greater than 95%. e) Determination of class and subclass of monoclonal antibodies: Class and subclass of monoclonal antibodies produced by cloned hybridoma cells are determined by known immunodiffusion techniques of Ouchterlony (agar-gel immunodiffusion method) using class and subclass specific rabbit antibodies ( Bionetics).

The subclasses of mAbs are as follows:

405B.33.3 : 7 i<

406A.23.7 : y2b<

407A.15.27 : 12aK

f) Enzyme immunoassay (ELISA):

Microtiter plates are covered with 0.5 <p>g per well of a t-PA preparation (purity >95#) in 100 µl PBS. Free binding capacity of the plate is saturated with a buffer consisting of 0.256 gelatin in PBS containing 0.2% NaN3 (w/v), pH 7.4. 100 µl of probe containing the monoclonal antibodies 405B.33.3, 406A.23.7 and 407A.15.27, respectively, are incubated in the wells at 37°C for 2 hours. The plates are washed with PBS containing 0.0556 Tween 20, then incubated at 37°C for 2 hours with a phosphatase-conjugated rabbit anti-mouse immunoglobulin preparation. The fixed enzyme is further developed by incubation (37°C, 30 to 60 min.) with a solution of the enzyme substrate p-nitroph enyl phosphate (1 mg/ml in diethanolamine buffer 1056 containing 0.5 mM MgCl2 and 0.0256 (w/ v) NaN3, pH 9.8) and measurement of optical density at 405 nm.

The same ELISA was also performed using urokinase. Neither mAb binds to urokinase. All mAbs are t-PA specific.

g) Casein lysis assay (neutralization test):

To determine the inhibitory effect of the mAb, t-PA is first mixed with the mAbs 405B.33.3, 406B.23.7 and 407A.15.27, respectively, and incubated for 30-60 min. at 4°C and then the usual casein/plasminogen agar assay is performed (see Example 20B). None of the mAbs inhibit t-PA activity with the exception that mAb 405B.33.3 causes a delay (more than 6 hours) of casein lysis.

Example 22: Purification of hybrid plasminogen activator. general procedure

Extracts from transformed yeast cells are prepared as described in Example 16. Extracts from plasmid-transformed mammalian cells, such as CHO cells, are prepared as follows: The cells are first grown to 70-8056 density. The cell monolayer is then rinsed with medium as described above, with the exception that the addition of serum is omitted followed by cultivation of the cells for an additional period of 5-7 days. The medium is harvested every 24 h, at the same time fresh medium is supplemented to the cells. The conditioned medium thus obtained is then centrifuged at 5000 x g for 30 min. and filtered through a 0.45 pm filter to remove unwanted cell debris before affinity chromatography. Either immobilized protease inhibitor DE-3 from Erythrina latissima or immobilized antibodies to u-PA or t-PA are used as affinity matrix.

Hybrid PAs containing the catalytic B chain of t-PA are purified from the conditioned medium prepared as described above or from yeast cell extracts using the protocol originally developed for the purification of t-PA from melanoma cell-conditioned medium [cf. C. Heussen et al., J. Biol. Chem. 259, 11635-11638 (1984)].

All hybrid PAs are purified using polyclonal antibodies raised in rabbit or goat against parental u-PA and t-PA enzymes or using monoclonal antibodies (of mouse origin) raised against the parental enzymes detected an epitope present in it hybrid PA as applicable (cf. example 21). The selected antibody is immobilized on a non-soluble matrix such as Affigel or Sepharose-4B. The conditioned medium prepared as described above or the yeast cell extract is then applied to an affinity matrix column, where unwanted proteins are washed away using an appropriate buffer, for example Dulbecco's PBS [0.1 g/l CaCl2, 0.2 g/l KC1, 0.2 g/l KH2PO4 0.047 g/l MgCl2, 8.0 g/l NaCl, 1.15 g/l Na2HP04; J. Exp. Med. 99, 167 (1954)] and then PA is eluted from the column using the chaotropic agent potassium thiocyanate [cf. M. Einarsson et al., Biochim. Biophys. Acta 830, 1-10 (1985)] or a low pH buffer such as 0.1-0.2 M glycine-HCl (pH 2.1).

After purification with monoclonal antibodies, hybrid PA has a purity of more than 90%.

Example 23: Purification of UK2UPA<B>(BC)

a. Preparation of a DE-3 Sepharose® column.

Per ml of cyanogen-bromide-activated Sepharose 4B® (Pharmacia) 5 mg of purified inhibitor from Erythrina latissima [F.J. Joubert et al., Hoppe-Seyler's Zeitschr. Physiol.Chem. 302, 531 (1981)], according to the manufacturer's instructions. The matrix is equilibrated with 0.2 M ammonium acetate buffer pH 7.0 containing 0.2 M NaCl, 0.1$ Synperionic® and 0.02$ sodium azide.

b. Chromatographic purification of UK2UPA<B>(BC) on DE-3 Sepharose®

The conditioned medium (cf. Example 22) is made 0.1% with respect to Synperonic® and then added to DE-3 Sepharose®. After gentle stirring for 1 hour at 4°C, DE-3 Sepharose 4B® is poured into a column and washed with 0.2 M NaCl, 0.1% Synperonic until the UV absorbance at 280 nm reaches the baseline level indicating no some remaining proteins in the eluate. Washing continues with 0.2 M ammonium acetate buffer pH 7.0 containing 0.2 M ammonium thiocyanate and 0.1% Synperonic. After the UV absorbance at 280 nm indicates that no more protein is present in the eluate, the column is eluted with 0.2 M ammonium acetate buffer pH 7.0 containing 1.6 M ammonium thiocyanate and 0.1% Synperonic®. Fractions containing the highest amidolytic activities, as measured by the fluorometric assay with Cbz-Gly-Gly-Arg-AMC as substrate [M. Zimmermann et al., Proe.Nati.Acad.Sei. USA. 75, 750 (1978)], are merged. At least 8056 of the activity applied to DE-3 Sepharose 4B® material is recovered as a single peak.

The pooled active fractions are dialyzed against 0.2 M ammonium acetate buffer pH 7.0 containing 0.1% Synperonic® and applied to a column containing monoclonal antibody 407A.15.27 directed against the first pretzel domain of t-PA, and coupled to Sepharose 4B® , equilibrated in 0.2 M ammonium acetate buffer pH 7.0 containing 0.156 Synperonic®, to remove endogenous t-PA. The eluate, which contains UK2UPA<B>(BC), is collected.

Reverse phase HPLC of purified UK2UPA<B>(BC) on a Nucleosil® 300-5-C18 column of dimensions 4 x 110 mm shows a single peak when eluting with a linear gradient over 30 min starting 7056 solution A consisting of water containing 0.156 trifluoroacetic acid and <3>056 solution B consisting of acetonitrile containing 0.0856 trifluoroacetic acid and ending at 4056 A and 6056 B. The purified proteins showed, by N-terminal sequence analysis of the first ten amino acid residues, the sequence SNELHOVPSN, which is identical to the the sequence expected from the DNA sequence that codes for the molecule.

Example 24: Purification of FK2UPA<B>(BC) and K2UPA<B>(BC)

a. Preparation of antibody affinity columns.

Rabbit anti-uPA antibodies purified from rabbit anti-uPA serum, monoclonal antibodies 405B.33.3 and 406A.23.7, are coupled to cyanogen bromide-activated Sepharose 4B® (Pharmacia) according to the manufacturer's instructions using 6 mg of antibody per ml activated Sepharose. The gel matrix is equilibrated with PBS containing 0.156 Synperonic® and 0.156 sodium azide.

b. Chromatographic purification of FK2UPA<B>(BC) and K2UPA<B>(BC) on antibody Sepharose 4B.

Conditioned medium (cf. Example 22) is made to 0.156 with regard to Synperonic® and added to anti-uPA Sepharose-4B or to 405B.33.3 or to 406A.23.7 Sepharose 4B. The last two antibodies are directed against the second kringle domain of t-PA. After gentle stirring for 2 hours at 4°C, antibody Sepharose is poured onto a column and washed with PBS containing 1 M NaCl and 0.1% Synperonic® until the UV absorbance at 280 nm indicates that no more protein is present in the eluate . The column is then eluted with 0.2 M glycine-HCl buffer pH 2.5. Fractions are collected in tubes containing a neutralizing amount of 1 M Tris. Fractions containing the highest amidolytic activities, measured by fluorometric analysis with Cbz-Gly-Gly-Arg-AMC as substrate [M. Zimmermann et al., Proe.Nati.Acad.Sei USA. 75, 750

(1978)], are collected.

Reverse phase HPLC of purified FK2UPA<B>(BC) and K2UPA<B>(BC) on a Nucleosil® 300-5-C18 column of dimensions 4 x 110 mm shows a single peak when eluting with a linear gradient over 30 min. starting with 7056 solution A consisting of water containing 0.156 trifluoroacetic acid and 3056 solution B consisting of acetonitrile containing 0.0856 trifluoroacetic acid and stopping at 4056 A and 6056 B. N-terminal sequence analysis of the first five residues of the purified proteins results in the sequence SYQGN for K2UPA<B>(BC) and SYQVI for FK2UPA<B>(BC), which is identical to the sequence expected from the DNA sequence encoding each molecule.

Example 25: Activity assessment of hybrid plasminogen activators in the presence and without fibrinogen fragments

Double rate analysis as described by Verheyen et al. [Thromb. Haemost. 48, 266 (1982)], based on the transfer of plasminogen to plasmin by means of plasminogen activator, followed by the reaction of plasmin with the chromogenic plasmin substrate H-D-valyl-L-leucyl-L-lysine nitroanilide dihydrochloride, is used. The analysis is performed in a micro-titer plate with 96 wells and with a Titertek® micro-titer plate reader. The wells contain 120-X pl 0.1 mol/l Tris/HCl buffer at pH 7.5 containing 0.156 Tween 80, 20 pl Glu-plasminogen at 1.3 pmol/1 in the above-mentioned Tris buffer, 100 pl plasmin substrate at 0.7 mmol/1 in Tris buffer, X pl samples at known concentrations (X corresponds to 10, 20, 40 and 60 pl, respectively), or urokinase standard with defined activity expressed in International Units, and 10 pl stimulator (fibrinogen fragments) at 3 mg/ml in distilled water or 10 pl of distilled water if the experiments are carried out without a simulator. The increase in light absorption divided by the square root of the incubation time is proportional to the plasminogen activator activity at a known concentration of activator and is expressed in International Units. High molecular weight urokinase with a defined activity expressed in International Units (American Diagnostics) has been used as a standard. Each plasminogen activator has been analyzed under equal conditions without and with fibrinogen fragments, respectively. Under these conditions, the differences in activities obtained become a measure of the stimulation of plasminogen activators by the fibrinogen fragments. Table 2 contains the results of the assay, which indicate no stimulation for the urokinase standard in contrast to the stimulation exerted by the fibrinogen fragments on the novel plasminogen activator molecules containing the catalytic domain of urokinase. Despite the absence of one or more of the non-catalytic domains F, G, K1 or K2 of tissue plasminogen activator, stimulation by the fibrinogen fragments is observed for all hybrid molecules tested.

Example 26: "Clot"-lysis activity of mutant plasminogen activators

Clot lysis activities are determined using the assay described by R.D. Philo and P.J. Gaffney [Thromb. Haemost. 45, 107-109 (1981)]. A logarithmic plot of lysis time against plasminogen activator concentration results in a straight line. The specific activity of a plasminogen activator is determined by comparison with the curves obtained in a standard preparation of tissue plasminogen activator or urokinase.

The curves of all activators measured have approximately the same slope allowing a direct relationship between the time required for clot lysis and their specific activity. As the different plasminogen activators do not have the same molecular weight, the specific activities must be expressed in a molar concentration, instead of the usual weight concentration, in order to obtain a meaningful criterion for the effectiveness of the different molecules. UK2TPA<B>(BC) is at least as active as standard t-PA, while FGK2UPA<B>(BC) and FK2UPA<B>(BC) show activities almost equal to t-PA but significantly higher than standard u-PA . K2UPA<B>(BC) has an activity almost equal to u-PA standard. The activities for the analysis are summarized in Table 3:

Example 27: Removal of plasminogen activator mutant molecules from the circulation of rabbits.

1. Marking.

All mutant molecules are labeled with 12E>j using the Iodogen method [P.J. Fraker et al., Biochem. Biophys. Res. Commun. 80, 849-857 (1978)].

To remove free <125>J that is in excess, the mutant molecules are affinity-purified either using the method described in Example 23 (PA having the t-PA B chain) or the method described in Example 24 (PA having u -PA B chain).

Specific radioactivities of 2-20 µgCi/µg proteins are usually obtained. The homogeneity of the labeled molecules is followed by SDS electrophoresis and then by X-ray autoradiography. In all cases, the mutant molecules migrate under non-reducing conditions as single bands and with Mv's identical to the untagged proteins.

2. Removal studies.

The experiments are carried out in New Zealand white rabbits weighing between 1.8 to 2.4 kg. The animals are anesthetized with 1750 mg/kg Urethan®

(Merck, Darmstadt, Germany) subcutaneously. Operative opening of the trachea is performed and a plastic tube is inserted into the external jugular vein and the common carotid artery. 0.5 ml of phosphate-buffered saline containing approximately 300-500 ng of mutant PA is injected into the jugular vein and serial blood samples (2 ml each) are drawn sequentially over a 60 min period. interval via the carotid artery.

The blood samples are taken on citrate, are centrifuged immediately at 3000 rpm for 15 min. and plasma is decanted. Aliquots are precipitated in 10% trichloroacetic acid and the precipitates are counted in a 7-counter.

Relative to t-PA isolated from the Bowes melanoma cell line, the mutant molecules show the following half-lives in circulation.

The pattern of removal of t-PA is bi-exponential with a very rapid a-phase followed by a slower p-phase elimination. Elimination of UK2TPA<B>(BC) and K2UPA<B>(BC) is almost monophasic, suggesting that distribution to another moiety is suppressed.

3. Organ distribution.

Rabbits are treated as above. 20 min. after the injection of the iodinated mutant molecules, the rabbits are sacrificed, and the major organs are removed, and the weight is determined and an aliquot is counted in the "Y-counter" after the homogenization.

Mutant PA has an even greater fraction of the radioactive molecules in circulation (supra), this suggests a reduced liver removal. The liver's reduced uptake is therefore the explanation for the extended half-life and the monophasic elimination pattern of the mutant molecules, especially UK2TPA<B>(BC) and K2TJPA<B>(BC).

In Examples 28-34, the plasmids pCGC5/K2UPA<B>, pCGC6/FTJPAB, pCGC7/FK2UPA<B> and pCGC8/FGK2UPA<B> (cf. Example 18) are used for the construction of yeast expression plasmids. The yeast invertase signal sequence is fused in-frame to different encoded sequences. They are expressed under the control of the inducible PHO5 promoter. In some constructs, the glycosylation site is mutated.

Example 28: Cloning of phage F1 origin of replication into expression vector pJDB207: Plasmids belonging to the pEMBL family [Dente et al., Nucl. Acids Res. 11, 1645-55 (1983)] contains a region of the genome of phage F1 with all cis-acting elements for DNA replication and morphogenesis. Only in superinfection with phage Fl (helper) are large amounts of single-stranded plasmid DNA secreted into the medium.

Plasmid pEMBL19(+) is cut with Seal and EcoRI. A 2.2 kb fragment is isolated containing parts of the ampicillin resistance gene of pBR322 (the Scal site), the Fl intergenic region and parts of the p<->galactosidase gene up to the polylinker region (the EcoRI site). Plasmid pJDB207 is linearized by cutting with HpaI. 10 µg of linearized plasmid is partially cut with 7.5 units of EcoRI in the presence of 0.1 mg/ml ethidium bromide for 40 min. at 37°C. The reaction is stopped by the addition of 10 mM EDTA. The 1.8 kb EcoRI-HpaI fragment is isolated on a preparative 0.8% agarose gel. 3 pg Hpal cut pJDB207 is further cut with Seal. The 4.8 kb HpaI-Scal fragment is isolated. The DNA fragments are electroeluted from the agarose gel blocks, purified by DE52 chromatography and ethanol precipitated.

0.2 pmol of each of the 2.2 kb Seal-EcoRI fragments and the 1.8 kb EcoRI-HpaI fragment and 0.1 pmol of the HpaI-Scal vector fragment are ligated. The ligation mixture is used to transform competent E.coli HB101 Ca<2+> cells.

Plasmid DNA of 12 ampicillin-resistant colonies is analyzed by EcoRI/Pstl double digestion. DNA from a single clone with correct restriction fragments is selected and referred to as pJDB207FlLac.

Example 29: Construction of plasmid pJDB207/PH05-I-FK2UPA<B>: The encoded sequence of FK2UPA<B> as it is in plasmid pCGC7/FK2UPA<B> is used for expression in yeast by fusing the yeast invertase signal sequence and expressing the gene under the control of the PHO5 promoter.

Plasmid pCGC7/FK2UPA<B> (see Example 18D) is cut with PstI and BamHI. The 1147 bp Pstl-BamHI fragment contains the FK2UPA<B> encoded sequence from the Pstl site at nucleotide position 199 of t-PA (Fig. 1) to the BamHI site at position 1322 of u-PA (Fig. 3).

Plasmid pJDB207/PH05-I-TPA (see Example 6C) is cut with SalI and Pstl. The 891 bp fragment is isolated. It contains the PHO5 promoter, the invertase signal sequence and 19 bases of t-PA (the Pstl site).

Plasmid pJDB207/PHO5-1-UPA (see Example 8) is cut with SalI and BamHI. The 6.6 kb vector fragment contains the 3' part of the u-PA gene from the BamHI site at nucleotide position 1323 (Fig. 3) to position 1441 (the PvuII site with Xhol linker added) and the PHO5 transcription termination signals.

0.2 pmol of each of the 891 bp SalI-Pstl fragment and the 1147 bp Pstl-BamHI fragment and 0.1 pmol of the 6.6 kb SalI-BamHI vector fragment are ligated and used to transform E.coli HB101 Ca<2+> cells. 8 ampicillin-resistant colonies are grown in LB medium containing ampicillin (100 mg/l). Plasmid DNA is isolated and analyzed by EcoRI and HindIII restriction digestion. A plasmid with the expected restriction fragments is selected and referred to pJDB207/PH05-I-FK2UPA<B>.

Plasmid pCGC6/FUPA<B> and pCGC8/FGK2UPA<B> can be used in the same way as pCGC7/FK2UPA<B>. The resulting yeast expression plasmids are designated pJDB207/ PH05-I-FUPAB and pJDB207/ PH05-I-FGK2UPA<B> respectively.

Example 30: Mutation of the glycosylation site at fAsn 3021 of the urokinase B chain: a) Cloning of a Pstl-BamHI fragment from u-PA into M13mpl8: The plasmid pJDE207/ PHO5-1 -UPA (see example 8) contains the entire encoded region to urokinase. DNA is cut with PstI and BamHI. The 886 bp Pstl-BamHI fragment from the urokinase gene contains the glycosylation site (Asn 302) at nucleotide positions 1033-1041. Another fragment of similar size is further cut with BstEII. The 886 bp Pstl-BamHI fragment is isolated on a preparative 0.856 agarose gel.

M13mpl8 RF DNA is cut with Pstl and BamHI. The 7.3 kb fragment is isolated on a preparative 0.856 agarose gel. The DNA fragments are electroeluted from the agarose gel and purified by DE52 chromatography and ethanol precipitation.

0.1 pmol of the 7.3 kb PstI-BamHI cut vector and 0.2 pmol of the 886 bp Pst-BamHI u-PA fragment are ligated. 1 µl and 3 µl of the ligation mixture are used for transformation of E.coli JM109 Ca<2+> cells according to the manual "M13 Cloning and Sequencing Manual" published by Amersham. 12 colorless plaques are picked and single-stranded DNA prepared [J. Messing, Methods in Enzymology 101, 21-78 (1983)]. The single-stranded DNA is used to prepare partially double-stranded DNA by base pairing and by extending the M13 universal primer with Klenow polymerase. The reaction product is extracted with phenol/chloroform and the DNA is precipitated with ethanol. DNA is cut with Pstl and BamHI. An 886 bp fragment indicates that the u-PA fragment has been cloned into the M13mpl8 vector. A clone is further analyzed and the correct insert is confirmed by sequencing. The clone is referred to as M13mpl8/UPA.

b) Mutation of the glycosylation site of Asn302:

The mutagenic and sequencing primers are synthesized using the phosphoramidite method [M.H. Caruthers, in: Chemical and Enzymatic Synthesis of Gene Fragments, (ed. H.G. Gassen and A. Lang) Verlag Chemie, Weinheim, Federal Republic of Germany] on an Applied Biosystem Model 380B synthesizer.

In vitro mutagenesis on single-stranded template DNA is performed as described by T.A. Kunkel [Proe. Nat. Acad. Pollock. USA 82, 488-492 (1985)]. Uracil containing single-stranded template DNA is produced during one growth cycle of the E.coli strain RZ1032 (dut~, young").

100 pmol of the mutagenic oligonucleotide primer W is phosphorylated in 20 µl of 50 mM Tris-HCl pH 7.5, 10 mM MgCl 2 , 5 mM DTT, 0.5 mM ATP and 20 units of T4 polynucleotide kinase (Boehringer). After 30 min. at 37°C the reaction is stopped by heating to 70°C for 10 min.

0.3 pmol of uracil containing M13mpl8/UPA template DNA is incubated with 10 pmol of phosphorylated mutagenic oligodeoxyribonucleotide primer ¥ and 10 pmol of M13 universal sequencing primer in 30 µl of 10 mM Tris-HCl pH 8.0, 10 mM MgCl2. The sample is heated to 80° C and left to cool at room temperature in a small water bath.

c) Extension-ligation reaction:

To the above base-paired sample is added 10 µl of an enzyme-dNTP mixture containing 1 mM dNTP, 10 mM Tris-HCl pH 8.0, 10 mM MgCl2, 20 mM DTT, 1 mM ATP, 400 units of T4 DNA ligase (Biolabs, 400 (U/pl) and 6 units of Klenow DNA- polymerase (Boehringer, 6 U/pl). Incubation is at 15°C overnight. d) Transformation of E.coli BMH71 cells: The ligation mixture is diluted to 200 µl with TE. 0.1 µl, 1 µl and 10 µl of the extension-ligation mixture are added to competent E.coli BMH71 Ca<2+> cells (Kunkel, supra). After 30 min. on ice, the cells are exposed to heat shock for 3 min. at 42°C and then stored on ice. The cells are plated with top agar and E.coli JM101 indicator cells. 6 plaques are picked and used for infection by E.coli JM109. The phages are isolated from the supernatant by PEG precipitation. Single-stranded DNA is prepared by extraction with phenol and precipitation with ethanol. Template DNA is resuspended in TE.

Mutation of the AAT codon (Asn302) to the CAA codon (Gln302) is confirmed in a clone by DNA sequencing with the above-mentioned sequencing primer using the chain termination method [F. Sanger et al., Proe. Nat. Acad. Pollock. USA 74, 5463-67 (1977)]. The mutation results in an Asn —> Gin change at amino acid position 302 to u-PA and thus eliminates the only glycosylation site in urokinase. W denotes the mutation of the glycosylation site in the u-PA B chain (Asn 302 —> Gln302). The positive clone is referred to as M13mpl8/UPA-W.

Example 31: Transfer of the mutation [Gln302] in the urokinase B chain to a FK2UPA<B> hybrid: Plasmid pJDB207/PE05-I-FK2UPA<B> is cut with SalI and Xhol. The 2.2 kb SalI-Xhol fragment is isolated, electroeluted from the agarose gel, purified by DE52 chromatography and precipitated in ethanol. This DNA fragment contains two Mstl sites in the PE05 promoter and the u-PA sequence. 3 pg of the 2.2 kb SalI-Xhol fragment is partially cut with 3 units of Mstl for 10 min. at 37°C. The reaction products are separated on a preparative 0.856 agarose gel and the 1651 bp SalI-Mstl fragment is isolated and electroeluted from the gel. The DNA fragment contains the SalI-BamHI sequence of pBR322, the PHO5 promoter, the invertase signal sequence and the FK2UPA<B> coded sequence up to the Mstl site in the u-PA part at nucleotide position 935.

RF DNA is prepared for M13mpl8/UPA-W (see Example 30) using the rapid DNA isolation procedure [D.S. Holmes et al., Analyt. Biochem. 114, 193-97 (1981)]. 5 pg of DNA is cut with BamHI and Mstl. After adding 2 pg RNase (Serva) and incubation for 5 min. at 37°C the 387 bp Mstl-BamHI fragment is isolated on a preparative 0.856 agarose gel. The DNA fragment is electroeluted and precipitated in ethanol. The fragment contains the mutation AAT —> CAA at nucleotide positions 1033-1035 (Asn302 —> Gin) in the urokinase B chain.

Plasmid <p>JDB207/PH05-I-UPA is cut with SalI and BamHI. The 6.6 kb vector fragment (see Example 29) is isolated. 0.2 pmol of the 1651 bp SalI-Mstl fragment, 0.2 pmol of the 387 bp Mstl-BamHI fragment and 0.1 pmol of the 6.6 kb SalI-BamHI vector fragment are ligated. Competent E.coli HB101 Ca<2+ >cells are transformed. 12 ampicillin-resistant transformants are grown. Plasmid DNA is isolated and analyzed by EcoRI and HindIII restriction digestion. The mutation (W) at the glycosylation site destroys the EcoRI site at nucleotide positions 1032-1037. The presence of the mutation is confirmed by DNA sequencing. A plasmid DNA with mutation in the u-PA B chain is referred to as pJDB207/PH05-I-FK2UPA<B->w. This plasmid has an intact glycosylation site in kringle K2, but the mutant site w(Asn302 —> Gin) in the u-PA B chain.

The plasmids pJDB207/ PH05-I-FUPAB-W and

pJDB207/PH05-I-FGK2UPA<B->W

are constructed in the same way starting from the corresponding non-mutated plasmids (see Example 29).

The 4.8 kb SalI-HpaI vector fragment of pJDB207/PH05-I-FK2UPAB-W is replaced by the 6.2 kb SalI-HpaI vector fragment of pJDB207/FlLac (see Example 28). The 6.2 kb fragment has a 1.4 kb FILac insert of pEML19 cloned into the 4.8 kb fragment of pJDB207. After ligation, transformation and analysis of the new construct, a correct plasmid with the FILac insert is referred to as

pJDB207/FlLac/PH05-I-FK2UPA<B->W.

Plasmid pJDB207FlLac/ PH05-1-FGKoUPAB-W is obtained in the same way.

Likewise, the 4.8 kb SalI-HpaI vector fragment of pJDB207/- PH05 -1 -MOU-KoTPAB (see Example 15C) is replaced by the 6.2 SalI-HpaI vector fragment of pJDB207FlLac. The resulting plasmid is referred to as PJDB207FlLac/PH05-I-UK2TPA<B>. The plasmids pJDB207FlLac/PH05-I-TJK2IJPAB, pJDB207FlLac/PH05-I-TPA<A>UPA<B> and pJDB207FlLac/PH05-I-UPA<A>TPA<B> are obtained in the same way from the plasmids without the FILac vector fragment.

Example 32: Mutation of the glycosylation site [Asnl84GlySer] in pretzel K2 to FK2UPA<B->W:

a) Preparation of single-chain template:

Plasmid pJDB207FlLac/ PH05-I-FKoUPAB-W is used to . transform competent E.coli RZ1032 Ca<2+> cells [T.A. Kunkel, supra]. An ampicillin resistant colony is grown in LB medium supplemented with 100 pg/ml ampicillin, 20 pg/ml thymidine and 20 pg/ml deoxyadenosine. At a cell density of 1*10<8>/ml, the cells are collected, washed in LB medium and resuspended in LB medium containing 100 µg/ml ampicillin and 0.25

>jg/ml uridine. At an OD600 of 0.3, excipient R408 (Pharmacia-PL Biochemicals, Inc.) is added at a m.o.i. at 20. The culture is shaken vigorously for 5 hours at 37°C. Uracil-containing single-stranded DNA in the medium is isolated as

described by T.A. Kunkel (supra). Starting from pEMBL19(+)

(see Example 28) the F1 region is cloned into pJDB207 in a counterclockwise orientation. The isolated single-stranded DNA is in the "sense" strand of the FKgUPA<2> insert in the expression plasmid.

b) Mutation of the glycosylation site in Asnl84 to kringle K2 of t-PA: The mutation concerns the third position of the consensus amino acid recognition sequence for glycosylation. Serl86 is replaced with Ala.

Protocol for the mutation is as described in example 30. Instead of the M13 universal sequencing primer, a PHO5 oligonucleotide primer with the formula 5'-AGTCGAGGTTTAGTATGGC-3' is used which hybridizes to nucleotides -60 to -77 from ATG in the PHO5 promoter. After the extension and the ligation reaction, competent E.coli BMH71 Ca<2+> cells [Kunkel, supra] are transformed. Ampicillin-resistant colonies are picked and grown in LB medium containing 100 mg/l ampicillin. Plasmid DNA is prepared and analyzed for the presence of the mutation by DNA sequencing. The mutation of the TCA codon to GCA results in a Ser —> Ala switch at amino acid position 186 of t-PA. The mutation in the third position in the consensus sequence eliminates the glycosylation site. A clone with the mutated DNA is referred to as pJDB207FlLac/PH05-I-FKoUPAB-WY.

Y designates the mutation of the glycosylation site at Asnl84 in K2 of t-PA and the W mutation at Asn302 of the u-PA B chain. The resulting non-glycosylated FK2UPA<B> hybrid protein has two amino acid changes: Serl86 —> Ala in the t-PA pretzel K2 and Asn302 —> Gin in the u-PA B chain.

The analogous mutation in plasmid pJDB207F^Lac/PE05-I-FGK2UPA<B->W (see Example 31) leads to plasmid pJDB307FlLac/ PH05-I-FKoUPAB-wY. which encodes a non-glycosylated FGK2UPA<B> hybrid protein.

Example 33: Construction of plasmid pJDB207/PH05-I-K2UPA<B->

WY:

The nucleotide sequence encoding the hybrid K2UPA<B>protein defined by the amino acid sequence tPA(Serl-Gln3)Glyl76-Arg275)-uPA(Ilel59-Leu411) is obtained in plasmid pCGC5/K2UPA<B.> For expression in yeast, the inducible PHO5 promoter used and the invertase signal sequence is fused in the correct reading frame to the K2UPA<B> coding region. Plasmid pCGC5/K2UPA<B> is cut with BglII and AccI. The 487 bp Bglll-AccI fragment is isolated. It contains the encoded sequence from the Bglll site of t-PA (nucleotide position 178) to the Accl site of u-PA (nucleotide position 779). The fragment is cut with HphI resulting in 4 fragments.

Two oligodeoxyribonucleotides with the following formulas

is synthesized using the phosphoramidite method on an Applied Biosystem Model 380B synthesizer. The oligonucleotides I and II form a double-stranded DNA linker. The 5 nucleotides at the overhanging 5' end are part of the yeast invertase signal sequence, followed by the t-PA encoded sequence (Serl-Gln3)(Glyl76-Thrl91) to the first HphI cut site at nucleotide position 752 (see Fig. 1 ). The glycosylation site at position 729-737 (AsnGLySer) is mutated in the synthetic sequence from TCA (Ser) to GCA (Ala), thus eliminating the glycosylation detection sequence. The mutation of the glycosylation site at amino acid positions 184-186 of t-PA (i.e., the second glycosylation site in native t-PA) is designated Y.

Oligonucleotides I and II are phosphorylated at their 5' ends, heated for 10 min. at 85°C and fused on cooling to room temperature. 10.5 pg (270 pmol) of kinase-treated, double-stranded linker DNA is ligated at a 30-fold molar excess to HphI cut DNA fragments (see above) as described in Example 8B. Linker molecules that are in excess are removed by precipitation with isopropanol. DNA is further cut with Seal. The 252 bp fragment is isolated on a preparative 1.556 agarose gel, electroeluted and precipitated in ethanol.

Plasmid p31RIT-12 (see Example 6B) is cut with SalI and XhoI. The isolated fragment is further cut with Hgal (see Example 6C) and BamHI. The resulting 591 bp BamHI-Hgal fragment is isolated. It contains the PHO5 promoter and invertase signal sequence.

Plasmid pJDB207/PH05-I-FK2UPAB-w is cut with BamHI. 5 pg of the linear DNA is partially cut with 10 units of Seal for 10 min. The reaction is stopped by the addition of 10 mM EDTA. The 7.7 kb BamHI-Scal vector fragment is isolated, electroeluted and precipitated in ethanol. It contains the 3' portion of the encoded sequence from the Scal site in t-PA (position 953) to the end of the u-PA B chain (the PvuII site at position 1441 with Xhol linker added), the PHO5 terminator and the pJDB207 vector sequences. 0.2 pmol of each of the 591 bp BamHI-Hgal fragment and the 252 bp sticky ends (linker)-ScaI fragment and 0.1 pmol of the 7.7 kb vector fragment are ligated. After transformation of E.coli HB101 Ca<2+> cells, 12 ampicillin resistant colonies are grown. Plasmid DNA is isolated and analyzed by EcoRI and HindIII cutting. Presence of the mutations is confirmed by DNA sequencing. A correct clone is selected and referred to as pJDB207/PH05-I-K2UPA<B->WY. The glycosylation sites in kringle K2 of the t-PA and u-PA B chain are both mutated (Y and W, respectively).

The resulting non-glycosylated K2UPA<B->hybrid protein has two amino acid changes: Serl86 —> Ala in the t-PA K2 domain and Asn302 —> Gin in the u-PA B chain.

Example 34: Mutation of the glycosylation sites [Asnl84GlySer] and [Asn 448ArgThr] in UK2TPA<B>.

Uracil containing single-chain template [T.A. Kunkel, supra] to plasmid pJDB207FlLac/PH05-I-UK2TPA<B> hybrid (see Example 31) is prepared as described in Example 30. The procedure for the mutation of the glycosylation site at Asnl84 is as described in Example 32. Mutation of the glycosylation site at Asn448 results in a Thr450 —> Ala amino acid change. Protocol for the mutation is described in Example 30. The phosphorylated mutagenic primers Y and Z are both base-paired to the uracil-containing single-chain template of pJDB207FlLac/PH05-I-IJK2TPAB. Additional use of the PE05 oligonucleotide primer (see Example 32) is optional.

After the extension and ligation reaction, competent E.coli BMH71 Ca<2+> cells are transformed. Plasmid DNA from ampicillin-resistant transformants is prepared and analyzed for the presence of both mutations by DNA sequencing with the indicated sequencing primers.

Plasmid DNA from a clone with both mutations is referred to as pJDB207FlLac/PH05-I-UK2TPA<]B->YZ. Y denotes the mutation to the glycosylation site at Asnl84 and Z the mutation at Asn448. The non-glycosylated UK2TPA<B> hybrid protein has two amino acid changes: Serl86 —> Ala in the K2 coil of t-PA and Thr450 —> Ala in the t-PA B chain.

The protocol for the mutation can also be used for templates of pJDB207FlLac/ PH05-I-UKoUPAB.

pJDB207FlLac/ PH05-I-TPAAUPAB and

pJDB207FlLac/ PH05-I-UPAATPAB (see Example 31)

with mutagen primer W for mutation of the glycosylation site in the u-PA B chain and/or mutagen primers Y and Z and others published in European patent application No. 225286 for mutation of the glycosylation sites in t-PA.

Example 35: Transformation of Saccharomyces cerevisiae GRF18 and preparation of yeast cell extracts.

The plasmids pJDB207/PH05-I-FK2UPA<B>,

pJDB207FlLac/ PH05-I-FK2UPAB-W, pJDB207FlLac/PH05-I-FK2UPA<B->WY, pJDB207FlLac/PH05-I-TJK2TPAB , pJDB207FlLac/PJ305-I-UK2TPAB-YZ,

PJDB2 0 7/ PHO 5-1-K2UPAB-WY. pJDB2 0 7/ PHO 5-1-FUPAB, pJDB 2 0 7/ PHO 5-1-FUPAB-W, pJDB207/PH05-I-FGK2UPA<B>, pJDB207/PH05-I-FGK2UPA<B->W, pJDB207FlLac /PH05-I-FGK2UPA<B->W and pJDB207FlLac/ PHO 5-1-FGKoUPAB-WY

is transformed into Saccharomyces cerevisiae strain GRF18 (DSM 3665). Transformation, cell growth and preparation of cell extracts are described in example 16.

The resulting hybrid plasminogen activators can be purified in a manner similar to that described in Examples 22 to 24.

Example 36: Preparation of freeze-dried hybrid plasminogen activators.

The solutions obtained in Examples 22 to 24 are further purified and freeze-dried as follows: The solution is diluted with 10 volumes of 0.1 M ammonium acetate pH 5.0 (total volume 80 ml) and applied to a column containing 5 ml of CM-Sepharose Fast Flow (Pharmacia) with an elution rate of 25 ml/h at room temperature. (The column has been pre-equilibrated with 0.1 M ammonium acetate). The product-free eluate is discarded. The column is washed with 15 ml of 0.1 M ammonium acetate pH 5.0 and with 10 ml of 0.1 M ammonium acetate pH 7.0. Elution of the adsorbed hybrid PA is then carried out at 1 M ammonium acetate pH 8.6 at room temperature (elution rate 5 ml/h). To prevent gas formation in the column, the elution is carried out at an excess pressure of 1 to 1.5 bar. The hybrid PA content in the eluate is measured with a UV monitor (280 nm). A fraction containing approximately 9056 of the eluted hybrid PA is collected and freeze-dried. The purity of the solid hybrid PA freeze-dried material is approximately or higher than 95% as assessed by HPLC. The product contains no detergents.

Example 37: First pharmaceutical preparation for parenteral administration.

A solution containing pure uPA(1-44)-tPA(176-527) obtained as described above is dialyzed against 0.3 molar sodium chloride containing 0.0156 Tween 80® and stored at -80°C. Before administration, the concentration is adjusted to 75 µg/ml of total PA and 0.3M NaCl. The solution is sterilized by filtration through a 0.22 pg membrane filter.

Instead of the above-mentioned PA, it is also possible to use the same amount of another PA described in the previous examples, such as, for example, uPA( 1-158 )-tPA(276-527), uPA(1-131) -tPA(263-527), tPA(1-275)-uPA(159-411), tPA(1-262)-uPA(132-411), uPA(1-44)-tPA(176-261)- uPA(134-411) , tPA(1-49)-tPA(262-275)-uPA(159-411), tPA(1-49 )-uPA(134-411 ) , tPA(1-49)-tPA (176-275 )-uPA(159-411),

tPA(1-49)-tpA(176-262)-uPA(132-411),

tPA(1-3)-tPA(176-275)-uPA(159-411),

tPA(1-86)-tPA(176-275)-uPA(159-411) or

tPA(1-86)-tPA(176-262)-uPA(132-411),

or a mutant hybrid PA, such as, for example, tPA(l-49)-tPA(262-275)-uPA(159-301), Gin, 303-411), tPA(l-49)-tPA(176 -185, Ala, 187-275)-uPA(159-301, Gn, 303-411),

uPA(1-44)-tPA(176-185, Ala, 187-449, Ala, 451-527),

tPA(1-3)-tPA(176-185, Ala, 187-275 )-uPA(159-301), Glin, 303-411) or

tPA(1-86)-tPA(176-185, Ala, 187-275 )-uPA(159-301), Glin, 303-411).

Example 38: Other pharmaceutical preparation for parenteral administration (dispersion for injection).

169.3 mg soybean lecithin (soybean phosphatide NC 95, producer: Nattermann, Cologne, Germany; purity 90-9696; content of fatty acids: linoleic acid 61-7156, linoleic acid 4-756, oleic acid 6-1356, palmitic acid 10-1556, stearic acid 1 .5-3.556) and 92.7 mg of pure sodium glycocholate are dissolved in 752.5 ml of sterilized water. The solution is adjusted to pH 7.4 with 1 N NaOE. 10 mg of freeze-dried uPA(1-44)-tPA(176-527) is added. The mixture is stirred until a clear solution appears. The solution is sterilized by filtering through a 0.22 pm membrane filter and filled into ampoules.

Instead of the above-mentioned PA, it is also possible to use the same amount of another PA described in the previous examples, such as, for example, uPA(1-158)-tPA(276-527), uPA(1-131 ) -tPA(263-527 ), tPA(1-275)-uPA(159-411), tPA(1-262)-uPA(132-411), uPA(1-44)-tPA(176-261)- uPA(134-411), tPA(1-49)-tPA(262-275)-uPA(159-411), tPA(1-49)-uPA(134-411), tPA(1-49 )-tPA (176-275)-uPA(159-411),

tPA(1-49 )-tpA(176-262)-uPA(132-411),

tPA(1-3 )-tPA(176-275)-uPA(159-411),

tPA(1-86 )-tPA(176-275)-uPA(159-411) or

tPA(1-86 )-tPA(176-262)-uPA(132-411),

or a mutant hybrid PA, such as, for example, tPA(l-49)-tPA(262-275)-uPA(159-301), Gin, 303-411), tPA(l-49)-tPA(176 -185, Ala, 187-275 )-uPA(159-301, Glin, 303-411),

uPA(l-44 )-tPA(176-185, Ala, 187-449, Ala, 451-527), tPA(l-3)-tPA(176-185, Ala, 187-275 )-uPA(159- 301 ), Gin, 303-411) or

tPA(1-86 )-tPA(176-185 , Ala, 187-275 )-uPA(159-301), Glin, 303-411).

Example 39: Third pharmaceutical preparation for parenteral administration (including large pill (bolus) injection).

100 mg of hybrid plasminogen activator or mutant hybrid plasminogen activator such as one of those mentioned in Examples 37 and 38 is dissolved in 1000 ml of 50 mM glutamic acid/sodium glutamate containing 0.7% NaCl, pH 4.5. The solution is filled in ampoules and can be used for intravenous (bolus) infusion.

Deposition of microorganisms

The following strains were deposited on October 23, 1987 at the "Deutsche Sammlung fur Mikroorganismen" (DSM), Grlsebachstrasse 8, D-3000 Gottingen (accession numbers given):

The following hybridoma cell lines were deposited November 20, 1987 at the "Collection Nationale de Cultures de Microorganismes", Institut Pasteur, Paris (CNCM) with the following accession numbers:

Claims (8)

1. Method for producing a single-chain hybrid plasminogen activator consisting of the following amino acid sequences of human t-PA shown in Figure 1 and human u-PA shown in Figure 3, where the hybrid plasminogen activator is selected from the group consisting of tPA(1-3)- tPA(176-275)-uPA(159-411), tPA(1-49)-tPA(176-275 )-uPA(159-411, and tPA(1-86 )-tPA(176-275)-uPA (159-411), characterized by growing under suitable nutrient conditions a transformed yeast or mammalian host cell containing a DNA coding for said hybrid plasminogen activator and isolating said hybrid plasminogen activator. 2. Method according to claim 1, characterized in that tPA(1-3)-tPA(176-275)-uPA(159-411) is produced. 3. DNA sequence, characterized in that it consists of the nucleotide sequences coding for a single-chain hybrid plasminogen activator consisting of the following amino acid sequences of human t-PA shown in Figure 1 and of human u-PA shown in Figure 3, the hybrid plasminogen activator being selected from the group consisting of tPA(1-3)-tPA(176-275)-uPA(159-411), tPA(1-49)-tPA(176-275)-uPA(159-411, and tPA(1-86) -tPA(176-275)-uPA(159-411). 4. DNA sequence according to claim 3, characterized in that it consists of the nucleotide sequences coding for tPA(1-3)-tPA(176-275)-uPA(159-411). 5 . Hybrid vector for use in a yeast or mammalian host cell, characterized in that it comprises a DNA sequence, coding for a single-chain hybrid plasminogen activator consisting of the following amino acid sequences of human t-PA shown in Figure 1 and of human u-PA shown in Figure 3, the hybrid plasminogen activator is selected from the group consisting of tPA(1-3)-tPA(176-275)-uPA(159-411), tPA(1-49)-tPA(176-275)-uPA(159-411, and tPA(1-86)-tPA(176-275 )-uPA(159-411), the hybrid vector being able to express the given gene. 6. Hybrid vector according to claim 5, characterized in that it comprises a DNA sequence coding for tPA(1-3)-tPA(176-275)-uPA(159-411). 7. Eukaryotic host cell transformed with a hybrid vector, characterized in that it comprises a DNA sequence coding for a single-chain hybrid plasminogen activator consisting of the following amino acid sequences of human t-PA shown in figure 1 and human u-PA shown in figure 3, the hybrid plasminogen activator being selected from the group consisting of tPA(1-3)-tPA(176-275)-uPA(159-411), tPA(1-49)-tPA(176-275)-uPA(159-411, and tPA(l -86)-tPA(176-275)-uPA(159-411), the host cell being able to express the given gene. 8. Eukaryotic host cell according to claim 7, characterized in that it is transformed with a hybrid vector comprising a DNA sequence coding for tPA(1-3)-tPA(176-275)-uPA(159-411).